Endor Labs

No entries match these filters. Clear filters

1. Mets Muse Wonder. evaluated against the Endor Labs Agent Security League | Endor Labs (2026-11-03) Nov 3, 2026 · issue 130
   • Updated detection Anti-cheating pipeline: recall-then-diverge detector
   • Updated detection Anti-cheating pipeline: shell file write visibility
   • New integration Muse Code (Meta) agent harness added to Agent Security League
2. announced Oct 29, 2026 · issue 127

   Endor Labs: One-click plugin in the Google Antigravity marketplace brings the Endor Labs MCPMCPModel Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. server into Antigravity so agents can check packages before a pull request.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

   └──▷ HOW TO FIND IT
   Trigger a full security review of your project inside an Antigravity session immediately after installing the plugin.
   📍Install the Endor Labs plugin from the Plugin marketplace in Antigravity, then prompt the Antigravity agent: 'Scan my project for security flaws, check all dependencies for vulnerabilities or malware, and flag any exposed secrets.'
   Delegate a vulnerable-dependency backlog item to the SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself. Remediation AURI Agent to get a safe upgrade applied and a pull request opened automatically.
   📍Inside Antigravity, ask the SCA Remediation agent: 'Fix the vulnerable dependency flagged in the latest scan.' The agent selects the safest upgrade using reachability and upgrade-impact analysis, applies it, adds tests, and opens a pull request with evidence in the PR body.
   • AURI Agents — SCA Remediation, AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. Triage, Dependency Decision Helper, and Vulnerability Explainer — are now available inside Antigravity, letting teams hand off backlog security work to pre-built agents grounded in Endor Labs reachability, upgrade-impact, findings, and policy data.
   • The SCA Remediation AURI Agent now selects the safest dependency upgrade using reachability and upgrade-impact analysis, applies the fix, adds tests, and opens a pull request with evidence in the PR body, running entirely inside an Antigravity session.

   2 details in Everything

3. Endor Labs + Google Antigravity: Secure code by default in agentic workflows | Endor Labs (2026-10-29) Oct 29, 2026 · issue 125

   Endor Labs adds an Antigravity Plugin Marketplace integration that brings the Endor Labs MCPMCPModel Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. server and AURI Agents into Google's agentic development platform with one-click⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

   └──▷ HOW TO FIND IT
   Trigger an automated dependency fix inside an Antigravity session — the SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself. Remediation agent selects the safest upgrade and opens a PR with evidence.
   📍Steps: In Google Antigravity, open the Plugin marketplace and install the Endor Labs plugin. Then in your Antigravity session, ask: 'Use the SCA Remediation agent to fix the vulnerable dependency in this project.' The agent applies reachability and upgrade impact analysis, opens a pull request, and attaches evidence to the PR body.
   Get a plain-language explanation of what a CVE means for your specific project before deciding whether to act on it.
   📍Steps: With the Endor Labs plugin installed in Antigravity, ask: 'Use the Vulnerability Explainer agent to explain CVE-2024-12345 in the context of this project.' The agent returns a project-specific impact summary and recommended remediation steps.
   • Install the Endor Labs plugin from Google Antigravity's Plugin marketplace with one-click setup to add the Endor Labs MCP server and AURI Agents to Antigravity-based agentic workflows.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
   • The Endor Labs MCP server now gives Antigravity agents real-time security context — checking packages for known vulnerabilities and malicious code before a dependency is added, scanning for secrets, and surfacing vulnerability details inline while code is written.
   • AURI Agents — SCA Remediation, AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. Triage, Dependency Decision Helper, and Vulnerability Explainer — are now available inside Antigravity, running on the developer's own infrastructure with their own LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. keys and source-control credentials.
   • The Dependency Decision Helper AURI Agent answers whether a package should be added before an agent installs it.
   • The Vulnerability Explainer AURI Agent turns a CVE ID into a plain-language explanation of impact and remediation steps specific to the project.

   5 details in Everything

4. Introducing AURI Fix for Open Source | Endor Labs (2026-10-27) Oct 27, 2026 · issue 123

   Endor Labs launches AURI Fix for Open Source, enabling code-level and config-level mitigations triggered via endorctl fix.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

   • Adds endorctl fix CLI command to trigger AURI Fix remediation workflows directly from the terminal.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
   • Introduces application-side code mitigations: AURI can change how an application calls a vulnerable dependency to break the exploit path, typically in fewer than 10 lines, without waiting for an upstream patch.
   • Introduces configuration-based mitigations: AURI identifies and sets the specific property or configuration flag that disables a vulnerable code path, re-evaluates the configuration to confirm no path remains, and opens a pull request.
   • Adds agentic remediation workflow: an AURI agent can triage findings, select the remediation method (upgrade, code mitigation, config change, patch, or replacement), apply the fix, and open a pull request autonomously.

   4 details in Everything

5. announced Oct 27, 2026 · issue 123

   Endor Labs: Adds endorctl fix CLI command to trigger AURI Fix directly from the command line, applying upgrades, code mitigations, config changes, or dependency patches as⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

   └──▷ TRY IT
   Trigger AURI Fix from the CLI to automatically assess a reachable finding and open a pull request with the safest remediation — upgrade, code mitigation, or config change.
   copy

   $ endorctl fix

   • Introduces AURI Fix for Open Source (private beta), enabling automated remediation of open source vulnerabilities through upgrades, first-party code mitigations, targeted dependency patches, or dependency replacement.
   • Adds configuration-based mitigation as a new remediation path — AURI can identify and set the specific config property that eliminates an exploitable path when the vulnerability is gated on a setting.
   • Supports policy-based and agentic remediation workflows: the Endor Labs policy engine can open a pull request automatically, or an AURI agent can triage, choose the method, apply the fix, and open the PR for human approval.

   3 details in Everything

6. Everything, everywhere, all at once… the week software broke | Endor Labs (2026-10-26) Oct 26, 2026 · issue 122
   • New vulnerability check eMudhra certificate authority distrust coverage
   • New vulnerability check PHP2Shell (CVE-2026-109618, CVE-2026-109619) container image detection
   • New detection LiteLLM 1.107.0 malicious package detection via Threat Center and Package Firewall
7. Everything, everywhere, all at once… the week software broke | Endor Labs (2026-10-24) Oct 24, 2026 · issue 119
   • New vulnerability check eMudhra CA distrust checks across language certificate bundles
   • New vulnerability check PHP2Shell (CVE-2026-109618, CVE-2026-109619) detection via container scanning
   • New detection LiteLLM 1.107.0 malicious release flagging via Threat Center and Package Firewall
8. Endor Labs + SpaceXSI: Securing every stage of agentic software delivery | Endor Labs (2026-10-22) Oct 22, 2026 · issue 122

   Endor Labs integrates with SpaceXSI Grok Build, adding AURI agent kit, coding agent governance hooks, and package firewall for agentic pipelines.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

   └──▷ TRY IT
   Bootstrap Endor Labs security scanning inside a Grok Build agent session without leaving the chat.
   copy

   $ /add-plugin endorlabs

   • Adds the Endor Labs Agent Kit for Grok Build, installable from the marketplace or via /add-plugin endorlabs in the agent chat, bringing AURI remediation agents—AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. Remediation and SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself. Remediation—into Grok Build sessions: the AI SAST Remediation agent proposes fixes with test plans for critical code flaws and can open a pull request on developer approval, while the SCA Remediation agent queries scan results, checks package usage, and applies upgrades, code changes, or Endor Patches, then rescans to confirm no new vulnerabilities were introduced; Agent Kit workflows can also run outside the IDE in CI or cloud runs through the Grok Build SDK.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
   • Adds Coding Agent Governance hooks that record every MCPMCPModel Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. tool call, shell command, and file operation and enforce policies before each action runs, with deterministic blocking that does not depend on model judgment.
   • Adds Endor Scores to rate each MCP server and skill for trust, mapped to the OWASPOWASPA nonprofit foundation that publishes open standards, guides, and ranked vulnerability lists—most famously the OWASP Top 10—giving cyber tools a vendor-neutral benchmark to test or report against. Top 10 for LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. Applications.
   • Adds Endor Labs Package Firewall integration that blocks confirmed malicious packages at the registry level for teams routing installs through it.
   • Integrates Endor Labs AI SAST into Grok Build sessions to scan first-party code for security flaws, trace data flow from untrusted input to vulnerable calls, and flag exposed secrets before commit, and integrates Endor Labs SCA to check direct and transitive dependencies, with reachability analysis reducing SCA findings by up to 92%.

   5 details in Everything

9. PHP2Shell: Critical vulnerabilities in PHP (CVE-2026-109618, CVE-2026-109619) | Endor Labs (2026-10-22) Oct 22, 2026 · issue 122
   • New vulnerability check PHP2Shell detection via container scanning (CVE-2026-109618, CVE-2026-109619)
10. Endor Labs + SpaceXSI: Securing every stage of agentic software delivery | Endor Labs (2026-10-21) Oct 21, 2026 · issue 116

    Endor Labs integrates with SpaceXSI's Grok Build to secure agentic coding via AURI, Agent Kit, Coding Agent Governance hooks, and Package Firewall.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    └──▷ TRY IT
    Bootstrap Endor Labs security scanning inside an active Grok Build agent session without leaving the chat interface.
    copy

    $ /add-plugin endorlabs

    • Introduces the Endor Labs Agent Kit for Grok Build, installable via the marketplace or by typing /add-plugin endorlabs in agent chat, enabling in-session scanning, triage, and remediation through specialized remediation agents (including an AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. Remediation agent and an SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself. Remediation agent) that can open pull requests for approved fixes, and supporting Agent Kit workflows outside the IDE — in CI or cloud runs — via the Grok Build SDK.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Adds Coding Agent Governance hooks that record every MCPMCPModel Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. tool call, shell command, and file operation and enforce security policies deterministically before each action runs — without depending on a model's judgment.
    • Adds Endor Scores for each MCP server and skill, mapped to the OWASPOWASPA nonprofit foundation that publishes open standards, guides, and ranked vulnerability lists—most famously the OWASP Top 10—giving cyber tools a vendor-neutral benchmark to test or report against. Top 10 for LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. Applications, giving security teams a trust rating for approved and unapproved agent tooling.
    • Adds an MCP server integration that checks for code security risks, exposed secrets, and vulnerable dependencies from within an agent session, and a hooks integration that scans every package an agent installs, blocking confirmed malicious packages at install time.
    • Extends Endor Labs Package Firewall protection to Grok Build, blocking malicious packages at the registry level for teams that route installs through it.

    +1 more show less

    • Integrates Endor Labs AI SAST into Grok Build to scan first-party code for security flaws, trace data flow from untrusted input to vulnerable calls, and flag exposed secrets before commit, and integrates Endor Labs SCA into Grok Build to check direct and transitive dependencies with reachability analysis and risk scoring, cutting SCA findings by up to 92%.
    • Introduces the Endor Labs Agent Kit for Grok Build, installable via the marketplace or by typing /add-plugin endorlabs in agent chat, enabling in-session scanning, triage, and remediation through specialized remediation agents (including an AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. Remediation agent and an SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself. Remediation agent) that can open pull requests for approved fixes, and supporting Agent Kit workflows outside the IDE — in CI or cloud runs — via the Grok Build SDK.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Adds Coding Agent Governance hooks that record every MCPMCPModel Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. tool call, shell command, and file operation and enforce security policies deterministically before each action runs — without depending on a model's judgment.
    • Adds Endor Scores for each MCP server and skill, mapped to the OWASPOWASPA nonprofit foundation that publishes open standards, guides, and ranked vulnerability lists—most famously the OWASP Top 10—giving cyber tools a vendor-neutral benchmark to test or report against. Top 10 for LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. Applications, giving security teams a trust rating for approved and unapproved agent tooling.
    • Adds an MCP server integration that checks for code security risks, exposed secrets, and vulnerable dependencies from within an agent session, and a hooks integration that scans every package an agent installs, blocking confirmed malicious packages at install time.
    • Integrates Endor Labs AI SAST into Grok Build to scan first-party code for security flaws, trace data flow from untrusted input to vulnerable calls, and flag exposed secrets before commit, and integrates Endor Labs SCA into Grok Build to check direct and transitive dependencies with reachability analysis and risk scoring, cutting SCA findings by up to 92%.

    1 more in Everything

11. PHP2Shell: Critical vulnerabilities in PHP (CVE-2026-109618, CVE-2026-109619) | Endor Labs (2026-10-21) Oct 21, 2026 · issue 116
    • New vulnerability check PHP2Shell: CVE-2026-109618 and CVE-2026-109619 detection via container scanning
12. Your model, our analysis: BYOK for Endor Labs AI SAST | Endor Labs (2026-10-20) Oct 20, 2026 · issue 115

    Endor Labs AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. adds BYOKBYOKBring Your Own Key — a cloud deployment model where the customer supplies and controls the encryption keys rather than the provider, so sensitive data stays under the customer's cryptographic authority. support routing model requests through customer-owned LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. provider accounts, with an AI Gateway enforcing encrypted key store, provider allowlist, and tenant isolation.

    • New supported platformAdds BYOK (bring your own key) support for AI SAST and AI Chat in the Endor Labs AURI platform, routing all model requests through your own OpenAI, Azure OpenAI, Gemini, or Vertex AI account instead of Endor Labs' credentials — with no fallback to Endor Labs if a key is configured.
    • Supports passing provider API keys as environment variables during AI SAST scans so keys are never stored with Endor Labs.
    • Adds an Endor Labs AI Gateway that enforces an encrypted key store, a provider allowlist, and tenant isolation for all BYOK model requests.
    • Adds custom organizational context ingestion for AI SAST — including Infrastructure as Code files, repository files such as AGENTS.md and CLAUDE.md, triage feedback, and uploaded threat models — to inform analysis with deployment and triage history.

    4 details in Everything

    • New supported platform C language support for AI SAST
13. announced Oct 16, 2026 · issue 113
    • New integration Container image scanning integrated into Endor Labs software supply chain platform
    • New integration Container provenance via artifact signing integrated into build and deploy stages
    • New compliance framework SBOM and VEX support for container scanning compliance obligations
14. announced Oct 16, 2026 · issue 113

    Endor Labs expands its SSCS platform with CI/CD Security: pipeline discovery, RSPM, and build integrity verification via artifact signing.

    • Adds Pipeline Discovery to map tools and workflows across all repositories, surfacing shadow engineering, security coverage gaps, and policy violations.
    • Adds Build Integrity Verification via artifact signing, attaching provenanceprovenanceEvidence recording where a software artifact came from and how it was built, including its source and build process. It helps a consumer check that the artifact matches a trusted build. information (source repository, branch, commit, and repository owner) to packages and containers to confirm they were produced by CI/CD pipelines and have not been tampered with.

    2 details in Everything

    • New detection Repository Security Posture Management (RSPM) out-of-the-box policies
    • New compliance framework CIS benchmark and SLSA/NIST SSDF compliance coverage via RSPM
15. Container Scanning + SCA = Better Together | Endor Labs (2026-10-16) Oct 16, 2026 · issue 110

    Endor Labs adds container image signing at build time, signature verification at deploy time, and unified cross-dependency PR comments for vulnerability findings.

    • New vulnerability checkAdds container image signing at build time and signature verification at deploy time, with scan results consumable by CNAPPCNAPPA cloud-native application protection platform that combines checks across cloud configuration, workloads and development. It helps teams connect risks that separate tools might report apart. and admission controllers.
    • Includes container vulnerability findings in pull-request comments alongside application dependency findings, so developers see all impacted packages in one place.
    • New integrationEnables tracing of a production container finding back to its source repo, release artifact, and build via integration with runtime CNAPP tools.

    3 details in Everything

16. Introducing CI/CD Security with Endor Labs | Endor Labs (2026-10-16) Oct 16, 2026 · issue 110

    Endor Labs expands its SSCS platform with CI/CD security: pipeline discovery, repository posture management, and artifact signing with provenanceprovenanceEvidence recording where a software artifact came from and how it was built, including its source and build process. It helps a consumer check that the artifact matches a trusted build..

    • New integrationAdds Pipeline Discovery to map every tool touching code across repositories and CI/CD workflows, surfacing shadow engineering, security coverage gaps, and policy violations.
    • New compliance frameworkAdds Build Integrity Verification via artifact signing, attaching provenance information (source repository, branch, commit, repository owner) to signed packages and containers.

    2 details in Everything

    • New detection Repository Security Posture Management (RSPM) out-of-the-box policies
17. Introducing CI/CD Security with Endor Labs | Endor Labs (2026-10-16) Oct 16, 2026 · issue 016

    Endor Labs expands its SSCS platform with CI/CD security: pipeline discovery, repository posture management, and artifact signing with provenanceprovenanceEvidence recording where a software artifact came from and how it was built, including its source and build process. It helps a consumer check that the artifact matches a trusted build..

    • New integrationAdds Pipeline Discovery to map every tool touching code across repositories and CI/CD workflows, surfacing shadow engineering, security coverage gaps, and policy violations.
    • New compliance frameworkAdds Build Integrity Verification via artifact signing, attaching provenance information (source repository, branch, commit, repository owner) to signed packages and containers.

    2 details in Everything

    • New detection Repository Security Posture Management (RSPM) out-of-the-box policies
18. Container Scanning + SCA = Better Together | Endor Labs (2026-10-16) Oct 16, 2026 · issue 016

    Endor Labs adds container image signing at build time, signature verification at deploy time, and unified cross-dependency PR comments for vulnerability findings.

    • New vulnerability checkAdds container image signing at build time and signature verification at deploy time, with scan results consumable by CNAPPCNAPPA cloud-native application protection platform that combines checks across cloud configuration, workloads and development. It helps teams connect risks that separate tools might report apart. and admission controllers.
    • Includes container vulnerability findings in pull-request comments alongside application dependency findings, so developers see all impacted packages in one place.
    • New integrationEnables tracing of a production container finding back to its source repo, release artifact, and build via integration with runtime CNAPP tools.

    3 details in Everything

19. Endor Labs + SpaceXSI: Securing every stage of agentic software delivery | Endor Labs (2026-10-15) Oct 15, 2026 · issue 110

    Endor Labs integrates with SpaceXSI's Grok Build to secure agentic coding via AURI, Agent Kit, and Coding Agent Governance.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    └──▷ TRY IT
    Bootstrap Endor Labs security scanning inside an active Grok Build agent session without leaving the IDE.
    copy

    $ /add-plugin endorlabs

    • Install the Endor Labs Agent Kit for Grok Build via the marketplace or by typing /add-plugin endorlabs in the agent chat to bring AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment., SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself., and remediation agents into the IDE session.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Adds Coding Agent Governance hooks that record every MCPMCPModel Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. tool call, shell command, and file operation and enforce policies deterministically before each action runs — without relying on a model's judgment.
    • Endor Scores rate each MCP server and skill for trust, mapped to the OWASPOWASPA nonprofit foundation that publishes open standards, guides, and ranked vulnerability lists—most famously the OWASP Top 10—giving cyber tools a vendor-neutral benchmark to test or report against. Top 10 for LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. Applications, giving security teams a basis to approve or block agents and skills.
    • Adds the Endor Labs Agent Kit with specialized AURI remediation agents — including an AI SAST Remediation agent and an SCA Remediation agent — that pull findings from scans, propose fixes, and open pull requests with developer approval.
    • Agent Kit workflows can run outside the IDE in CI or cloud runs through the Grok Build SDK.

    +3 more show less

    • Integrates Coding Agent Governance hooks to block confirmed malicious packages at install time during agent-driven dependency installation.
    • AI SAST traces data flow from untrusted input to vulnerable calls in first-party code and flags exposed secrets before they are committed.
    • SCA Reachability analysis cuts findings by up to 92% by identifying which vulnerable dependencies the application actually calls.
    • Install the Endor Labs Agent Kit for Grok Build via the marketplace or by typing /add-plugin endorlabs in the agent chat to bring AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment., SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself., and remediation agents into the IDE session.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Adds Coding Agent Governance hooks that record every MCPMCPModel Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. tool call, shell command, and file operation and enforce policies deterministically before each action runs — without relying on a model's judgment.
    • Endor Scores rate each MCP server and skill for trust, mapped to the OWASPOWASPA nonprofit foundation that publishes open standards, guides, and ranked vulnerability lists—most famously the OWASP Top 10—giving cyber tools a vendor-neutral benchmark to test or report against. Top 10 for LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. Applications, giving security teams a basis to approve or block agents and skills.
    • Adds the Endor Labs Agent Kit with specialized AURI remediation agents — including an AI SAST Remediation agent and an SCA Remediation agent — that pull findings from scans, propose fixes, and open pull requests with developer approval.
    • Agent Kit workflows can run outside the IDE in CI or cloud runs through the Grok Build SDK.

    +2 more show less

    • Integrates Coding Agent Governance hooks to block confirmed malicious packages at install time during agent-driven dependency installation.
    • AI SAST traces data flow from untrusted input to vulnerable calls in first-party code and flags exposed secrets before they are committed.

    1 more in Everything

20. Endor Labs + SpaceXSI: Securing every stage of agentic software delivery | Endor Labs (2026-10-15) Oct 15, 2026 · issue 016

    Endor Labs integrates with SpaceXSI's Grok Build to secure agentic coding via AURI, Agent Kit, and Coding Agent Governance.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    └──▷ TRY IT
    Bootstrap Endor Labs security scanning inside an active Grok Build agent session without leaving the IDE.
    copy

    $ /add-plugin endorlabs

    • Install the Endor Labs Agent Kit for Grok Build via the marketplace or by typing /add-plugin endorlabs in the agent chat to bring AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment., SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself., and remediation agents into the IDE session.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Adds Coding Agent Governance hooks that record every MCPMCPModel Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. tool call, shell command, and file operation and enforce policies deterministically before each action runs — without relying on a model's judgment.
    • Endor Scores rate each MCP server and skill for trust, mapped to the OWASPOWASPA nonprofit foundation that publishes open standards, guides, and ranked vulnerability lists—most famously the OWASP Top 10—giving cyber tools a vendor-neutral benchmark to test or report against. Top 10 for LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. Applications, giving security teams a basis to approve or block agents and skills.
    • Adds the Endor Labs Agent Kit with specialized AURI remediation agents — including an AI SAST Remediation agent and an SCA Remediation agent — that pull findings from scans, propose fixes, and open pull requests with developer approval.
    • Agent Kit workflows can run outside the IDE in CI or cloud runs through the Grok Build SDK.

    +3 more show less

    • Integrates Coding Agent Governance hooks to block confirmed malicious packages at install time during agent-driven dependency installation.
    • AI SAST traces data flow from untrusted input to vulnerable calls in first-party code and flags exposed secrets before they are committed.
    • SCA Reachability analysis cuts findings by up to 92% by identifying which vulnerable dependencies the application actually calls.
    • Install the Endor Labs Agent Kit for Grok Build via the marketplace or by typing /add-plugin endorlabs in the agent chat to bring AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment., SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself., and remediation agents into the IDE session.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Adds Coding Agent Governance hooks that record every MCPMCPModel Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. tool call, shell command, and file operation and enforce policies deterministically before each action runs — without relying on a model's judgment.
    • Endor Scores rate each MCP server and skill for trust, mapped to the OWASPOWASPA nonprofit foundation that publishes open standards, guides, and ranked vulnerability lists—most famously the OWASP Top 10—giving cyber tools a vendor-neutral benchmark to test or report against. Top 10 for LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. Applications, giving security teams a basis to approve or block agents and skills.
    • Adds the Endor Labs Agent Kit with specialized AURI remediation agents — including an AI SAST Remediation agent and an SCA Remediation agent — that pull findings from scans, propose fixes, and open pull requests with developer approval.
    • Agent Kit workflows can run outside the IDE in CI or cloud runs through the Grok Build SDK.

    +2 more show less

    • Integrates Coding Agent Governance hooks to block confirmed malicious packages at install time during agent-driven dependency installation.
    • AI SAST traces data flow from untrusted input to vulnerable calls in first-party code and flags exposed secrets before they are committed.

    1 more in Everything

21. Your model, our analysis: BYOK for Endor Labs AI SAST | Endor Labs (2026-10-14) Oct 14, 2026 · issue 109

    BYOKBYOKBring Your Own Key — a cloud deployment model where the customer supplies and controls the encryption keys rather than the provider, so sensitive data stays under the customer's cryptographic authority. routes AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. and AI Chat model requests through your own LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. provider account with an encrypted key store, provider allowlist, and tenant isolation via the Endor Labs AI Gateway.

    • New integrationAdds bring your own key (BYOK) support for AI SAST and AI Chat, routing all model requests through your own OpenAI, Azure OpenAI, Gemini, or Vertex AI account instead of Endor Labs' provider accounts.
    • Routes BYOK requests through the Endor Labs AI Gateway with an encrypted key store, provider allowlist, and tenant isolation.
    • Supports passing provider API keys as environment variables during AI SAST scans so keys are never stored with Endor Labs.
    • Honors the region you have configured in Azure OpenAI or Vertex AI, so model requests run in your chosen data-residency region.
    • Adds custom organizational context ingestion from Infrastructure as Code files, repository files such as AGENTS.md and CLAUDE.md, triage feedback, and uploaded threat models to teach AI SAST how your application is deployed.

    5 details in Everything

    • New supported platform First-class C language support for AI SAST
22. announced Oct 14, 2026 · issue 109

    Endor Labs: Supports region selection for BYOKBYOKBring Your Own Key — a cloud deployment model where the customer supplies and controls the encryption keys rather than the provider, so sensitive data stays under the customer's cryptographic authority. when the provider allows it, including Azure OpenAI and Vertex AI.

23. Your model, our analysis: BYOK for Endor Labs AI SAST | Endor Labs (2026-10-14) Oct 14, 2026 · issue 016

    BYOKBYOKBring Your Own Key — a cloud deployment model where the customer supplies and controls the encryption keys rather than the provider, so sensitive data stays under the customer's cryptographic authority. routes AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. and AI Chat model requests through your own LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. provider account with an encrypted key store, provider allowlist, and tenant isolation via the Endor Labs AI Gateway.

    • New integrationAdds bring your own key (BYOK) support for AI SAST and AI Chat, routing all model requests through your own OpenAI, Azure OpenAI, Gemini, or Vertex AI account instead of Endor Labs' provider accounts.
    • Routes BYOK requests through the Endor Labs AI Gateway with an encrypted key store, provider allowlist, and tenant isolation.
    • Supports passing provider API keys as environment variables during AI SAST scans so keys are never stored with Endor Labs.
    • Honors the region you have configured in Azure OpenAI or Vertex AI, so model requests run in your chosen data-residency region.
    • Adds custom organizational context ingestion from Infrastructure as Code files, repository files such as AGENTS.md and CLAUDE.md, triage feedback, and uploaded threat models to teach AI SAST how your application is deployed.

    5 details in Everything

    • New supported platform First-class C language support for AI SAST
24. Learn how Endor Labs detects malicious VS Code extensions | Endor Labs (2026-10-13) Oct 13, 2026 · issue 108

    Endor Labs extends Threat Center detection and Package Firewall blocking to VS Code extensions, stopping malicious installs and poisoned updates before they reach developer w⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    • Extends Threat Center malicious-package detection to the Visual Studio Marketplace, covering VS Code extensions alongside npm, PyPI, Go, Maven, and NuGetNuGetMicrosoft's package manager for .NET, hosting a public registry of reusable libraries so tools can pull in dependencies without bundling or manually maintaining third-party code..⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Extends Package Firewall blocking to VS Code extensions, preventing unsafe versions from appearing in the VS Code marketplace and blocking unsafe extensions installed through the command line.
    • Adds three Package Firewall policy controls for VS Code extensions: Malware (block versions identified as malicious), Minimum package age (hold back new versions until vetted), and Exceptions (always-allow list for trusted extensions).
    • Each Package Firewall rule for VS Code extensions can be set to block or warn, letting teams start in monitoring mode and tighten policy over time.
    • Logs every VS Code extension install so security teams can track workstation activity.

    +1 more show less

    • Deploys via existing MDMMDMMobile device management, centrally configuring and managing enrolled devices. Administrators use it to apply settings, distribute applications and enforce device security policies. with no agent, wrapper, or certificate required on developer devices.
    • Extends Threat Center malicious-package detection to the Visual Studio Marketplace, covering VS Code extensions alongside npm, PyPI, Go, Maven, and NuGetNuGetMicrosoft's package manager for .NET, hosting a public registry of reusable libraries so tools can pull in dependencies without bundling or manually maintaining third-party code..⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Adds three Package Firewall policy controls for VS Code extensions: Malware (block versions identified as malicious), Minimum package age (hold back new versions until vetted), and Exceptions (always-allow list for trusted extensions).
    • Each Package Firewall rule for VS Code extensions can be set to block or warn, letting teams start in monitoring mode and tighten policy over time.
    • Logs every VS Code extension install so security teams can track workstation activity.
    • Deploys via existing MDMMDMMobile device management, centrally configuring and managing enrolled devices. Administrators use it to apply settings, distribute applications and enforce device security policies. with no agent, wrapper, or certificate required on developer devices.

    1 more in Everything

25. Learn how Endor Labs detects malicious VS Code extensions | Endor Labs (2026-10-13) Oct 13, 2026 · issue 016

    Endor Labs extends Threat Center detection and Package Firewall blocking to VS Code extensions, stopping malicious installs and poisoned updates before they reach developer w⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    • Extends Threat Center malicious-package detection to the Visual Studio Marketplace, covering VS Code extensions alongside npm, PyPI, Go, Maven, and NuGetNuGetMicrosoft's package manager for .NET, hosting a public registry of reusable libraries so tools can pull in dependencies without bundling or manually maintaining third-party code..⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Extends Package Firewall blocking to VS Code extensions, preventing unsafe versions from appearing in the VS Code marketplace and blocking unsafe extensions installed through the command line.
    • Adds three Package Firewall policy controls for VS Code extensions: Malware (block versions identified as malicious), Minimum package age (hold back new versions until vetted), and Exceptions (always-allow list for trusted extensions).
    • Each Package Firewall rule for VS Code extensions can be set to block or warn, letting teams start in monitoring mode and tighten policy over time.
    • Logs every VS Code extension install so security teams can track workstation activity.

    +1 more show less

    • Deploys via existing MDMMDMMobile device management, centrally configuring and managing enrolled devices. Administrators use it to apply settings, distribute applications and enforce device security policies. with no agent, wrapper, or certificate required on developer devices.
    • Extends Threat Center malicious-package detection to the Visual Studio Marketplace, covering VS Code extensions alongside npm, PyPI, Go, Maven, and NuGetNuGetMicrosoft's package manager for .NET, hosting a public registry of reusable libraries so tools can pull in dependencies without bundling or manually maintaining third-party code..⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Adds three Package Firewall policy controls for VS Code extensions: Malware (block versions identified as malicious), Minimum package age (hold back new versions until vetted), and Exceptions (always-allow list for trusted extensions).
    • Each Package Firewall rule for VS Code extensions can be set to block or warn, letting teams start in monitoring mode and tighten policy over time.
    • Logs every VS Code extension install so security teams can track workstation activity.
    • Deploys via existing MDMMDMMobile device management, centrally configuring and managing enrolled devices. Administrators use it to apply settings, distribute applications and enforce device security policies. with no agent, wrapper, or certificate required on developer devices.

    1 more in Everything

26. Endor Labs and Bright Security Partner to Close the Gap Between Discovery and Remediation | Endor Labs (2026-10-12) Oct 12, 2026 · issue 106

    Endor Labs partners with Bright Security to combine reachability-backed static analysis with DASTDASTDynamic application security testing, probing a running application from the outside for vulnerabilities. It checks behavior exposed through the application's interfaces. runtime exploit validation via SARIFSARIFStatic Analysis Results Interchange Format, a standard for sharing code-analysis findings, including locations and rule details. It lets scanners report results to a common review interface. handoff.

    • New integrationReachability-narrowed dynamic testing: Endor Labs narrows findings to reachable vulnerabilities, then hands a SARIF file to Bright Security (Bright STAR) for runtime exploitability confirmation.
    • Runtime-confirmed prioritization: findings confirmed exploitable by Bright Security at runtime are automatically elevated, backed by both static and live proof-of-concept evidence.
    • CI gate workflow: Endor SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. + Bright STAR used as a CI gate to catch and prove exploitability before code reaches the production branch.
    • Lower token and compute costs for AI-driven remediation: verified, machine-readable evidence from both platforms reduces redundant reasoning by AI coding and remediation agents.

    4 details in Everything

27. Open source carries the world. Patching it at Mythos-scale can't fall to maintainers alone. | Endor Labs (2026-10-12) Oct 12, 2026 · issue 106

    Endor Labs launches Zero-Day Patches — AI-generated, verified fixes for novel open source vulns delivered in under 24 hours.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    • Joins the Linux Foundation's Project Akrites as a founding member, routing coordinated vulnerability disclosure and remediation through a shared channel alongside Anthropic, OpenAI, Amazon, Chainguard, Cisco, Citi, IBM, JPMorgan Chase, and Microsoft.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Applies function-level reachability analysis across direct and transitive dependencies to rank which zero-dayzero-dayA vulnerability for which defenders have had no prior opportunity to apply a fix, often because it was previously unknown. A zero-day exploit takes advantage of that gap. findings are actually reachable from a customer's specific application code.
    • Joins the Linux Foundation's Project Akrites as a founding member, routing coordinated vulnerability disclosure and remediation through a shared channel alongside Anthropic, OpenAI, Amazon, Chainguard, Cisco, Citi, IBM, JPMorgan Chase, and Microsoft.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    1 more in Everything

28. Why Endor Labs AI SAST for C finds what other tools miss | Endor Labs (2026-10-12) Oct 12, 2026 · issue 106

    Endor Labs AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. adds full C language support with buildless scanning in the IDE and on PRs.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    • Pairs AI SAST for C with C SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself. in the same platform, covering first-party C code and open-source dependencies together under a single analysis run.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Integrates C scanning into AURI by Endor Labs, providing an independent security harness for agentic SDLCSDLCThe Software Development Life Cycle: a structured process covering planning, design, coding, testing, and deployment. Cyber tools integrate into it to catch vulnerabilities before code ships rather than after. workflows with function-level call paths, reproducible evidence, and policy enforcement across any agent, model, or CI stage.
    • Pairs AI SAST for C with C SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself. in the same platform, covering first-party C code and open-source dependencies together under a single analysis run.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    1 more in Everything

29. Open source carries the world. Patching it at Mythos-scale can't fall to maintainers alone. | Endor Labs (2026-10-12) Oct 12, 2026 · issue 016

    Endor Labs launches Zero-Day Patches — AI-generated, verified fixes for novel open source vulns delivered in under 24 hours.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    • Joins the Linux Foundation's Project Akrites as a founding member, routing coordinated vulnerability disclosure and remediation through a shared channel alongside Anthropic, OpenAI, Amazon, Chainguard, Cisco, Citi, IBM, JPMorgan Chase, and Microsoft.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Applies function-level reachability analysis across direct and transitive dependencies to rank which zero-dayzero-dayA vulnerability for which defenders have had no prior opportunity to apply a fix, often because it was previously unknown. A zero-day exploit takes advantage of that gap. findings are actually reachable from a customer's specific application code.
    • Joins the Linux Foundation's Project Akrites as a founding member, routing coordinated vulnerability disclosure and remediation through a shared channel alongside Anthropic, OpenAI, Amazon, Chainguard, Cisco, Citi, IBM, JPMorgan Chase, and Microsoft.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    1 more in Everything

30. Why Endor Labs AI SAST for C finds what other tools miss | Endor Labs (2026-10-12) Oct 12, 2026 · issue 016

    Endor Labs AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. adds full C language support with buildless scanning in the IDE and on PRs.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    • Pairs AI SAST for C with C SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself. in the same platform, covering first-party C code and open-source dependencies together under a single analysis run.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Integrates C scanning into AURI by Endor Labs, providing an independent security harness for agentic SDLCSDLCThe Software Development Life Cycle: a structured process covering planning, design, coding, testing, and deployment. Cyber tools integrate into it to catch vulnerabilities before code ships rather than after. workflows with function-level call paths, reproducible evidence, and policy enforcement across any agent, model, or CI stage.
    • Pairs AI SAST for C with C SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself. in the same platform, covering first-party C code and open-source dependencies together under a single analysis run.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    1 more in Everything

31. Endor Labs and Bright Security Partner to Close the Gap Between Discovery and Remediation | Blog | Endor Labs (2026-10-12) Oct 12, 2026 · issue 105

    Endor Labs and Bright Security integrate reachability SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. with runtime DASTDASTDynamic application security testing, probing a running application from the outside for vulnerabilities. It checks behavior exposed through the application's interfaces. via SARIFSARIFStatic Analysis Results Interchange Format, a standard for sharing code-analysis findings, including locations and rule details. It lets scanners report results to a common review interface. handoff, enabling a CI gate and AI-agent-ready evidence for confirmed-exploitable findings only.

    • New integrationAdds a SARIF-based integration pipeline where Endor Labs' function-level reachability analysis passes a narrowed candidate list to Bright STAR for runtime exploit confirmation, replacing full-surface DAST sweeps.
    • Introduces Endor SAST + Bright STAR as a combined CI gate, catching reachable and runtime-confirmed exploitable code before it reaches the production branch.
    • Enables runtime-confirmed prioritization: findings Bright Security confirms exploitable at runtime are automatically elevated above reachable-but-unproven and unreachable findings.
    • Delivers machine-readable, verified evidence (call path from Endor Labs plus confirmed exploit path from Bright Security) to AI remediation agents, reducing token overhead for reachability and exploitability re-derivation.

    4 details in Everything

32. Endor Labs and Bright Security Partner to Close the Gap Between Discovery and Remediation | Blog | Endor Labs (2026-10-12) Oct 12, 2026 · issue 016

    Endor Labs and Bright Security integrate reachability SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. with runtime DASTDASTDynamic application security testing, probing a running application from the outside for vulnerabilities. It checks behavior exposed through the application's interfaces. via SARIFSARIFStatic Analysis Results Interchange Format, a standard for sharing code-analysis findings, including locations and rule details. It lets scanners report results to a common review interface. handoff, enabling a CI gate and AI-agent-ready evidence for confirmed-exploitable findings only.

    • New integrationAdds a SARIF-based integration pipeline where Endor Labs' function-level reachability analysis passes a narrowed candidate list to Bright STAR for runtime exploit confirmation, replacing full-surface DAST sweeps.
    • Introduces Endor SAST + Bright STAR as a combined CI gate, catching reachable and runtime-confirmed exploitable code before it reaches the production branch.
    • Enables runtime-confirmed prioritization: findings Bright Security confirms exploitable at runtime are automatically elevated above reachable-but-unproven and unreachable findings.
    • Delivers machine-readable, verified evidence (call path from Endor Labs plus confirmed exploit path from Bright Security) to AI remediation agents, reducing token overhead for reachability and exploitability re-derivation.

    4 details in Everything

33. Introducing AI Security Code Review | Blog | Endor Labs (2026-10-12) Oct 12, 2026 · issue 105

    Endor Labs launches AI Security Code Review — multi-agent PR analysis catching auth gaps, CSRFCSRFCross-site request forgery, an attack that makes a signed-in browser send an unwanted request to a site. The site may accept the request using the victim's existing session. breaks, and crypto misuse missed by SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment..⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    └──▷ HOW TO FIND IT
    Enable AI Security Code Review for your GitHub organization so every new PR is automatically analyzed for architectural security risks.
    📍In the Endor Labs console, go to your project settings and enable the Endor Labs GitHub App for your GitHub organization. Once enabled, AI Security Code Review will automatically run on all new pull requests and surface findings in the Security Review section of each project.
    • Adds AI Security Code Review, a new capability that deploys multiple AI agents (Developer, Architect, Security) to analyze every pull request for design flaws and architectural security changes missed by rule-based SAST/SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself. tools.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Detects security-impactful PR categories including: API endpoint exposure, access control changes, authentication modifications, CI/CD workflow additions, cryptographic library swaps, database schema changes, dependency privilege escalationprivilege escalationGaining permissions beyond those an account or process initially had. It can turn limited access into administrator control or access to another user's resources., input validation gaps, memory management changes, network security control modifications, payment processing integrations, and PIIPIIPersonally Identifiable Information — data that can identify a specific individual, such as names, email addresses, or SSNs. Cyber tools flag or redact it to limit exposure and meet privacy regulations. data handling.
    • Assigns each detected change a priority level (Critical, High, Medium, Low) based on context and potential impact, and surfaces findings in a dedicated Security Review section within each project.
    • Integrates with the Endor Labs GitHub App to automatically trigger analysis on new PRs across a GitHub organization, and surfaces submitter and reviewer identity in each finding.
    • Connects AI Security Code Review to Endor Labs' policy engine and existing SCMSCMSource Code Management: version-control systems (Git, SVN, etc.) that track changes to code over time, giving cyber tools reproducible builds, audit trails, and collaborative development workflows. and project management systems to automate ticket creation and routing — for example, sending compliance-related PII findings to GRCGRCGovernance, Risk, and Compliance — a framework for aligning IT controls with business policy, regulatory requirements, and risk management so organizations can audit and report on their posture systematically. teams or infrastructure changes to DevOps.

    5 details in Everything

34. Introducing AI Security Code Review | Blog | Endor Labs (2026-10-12) Oct 12, 2026 · issue 016

    Endor Labs launches AI Security Code Review — multi-agent PR analysis catching auth gaps, CSRFCSRFCross-site request forgery, an attack that makes a signed-in browser send an unwanted request to a site. The site may accept the request using the victim's existing session. breaks, and crypto misuse missed by SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment..⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    └──▷ HOW TO FIND IT
    Enable AI Security Code Review for your GitHub organization so every new PR is automatically analyzed for architectural security risks.
    📍In the Endor Labs console, go to your project settings and enable the Endor Labs GitHub App for your GitHub organization. Once enabled, AI Security Code Review will automatically run on all new pull requests and surface findings in the Security Review section of each project.
    • Adds AI Security Code Review, a new capability that deploys multiple AI agents (Developer, Architect, Security) to analyze every pull request for design flaws and architectural security changes missed by rule-based SAST/SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself. tools.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Detects security-impactful PR categories including: API endpoint exposure, access control changes, authentication modifications, CI/CD workflow additions, cryptographic library swaps, database schema changes, dependency privilege escalationprivilege escalationGaining permissions beyond those an account or process initially had. It can turn limited access into administrator control or access to another user's resources., input validation gaps, memory management changes, network security control modifications, payment processing integrations, and PIIPIIPersonally Identifiable Information — data that can identify a specific individual, such as names, email addresses, or SSNs. Cyber tools flag or redact it to limit exposure and meet privacy regulations. data handling.
    • Assigns each detected change a priority level (Critical, High, Medium, Low) based on context and potential impact, and surfaces findings in a dedicated Security Review section within each project.
    • Integrates with the Endor Labs GitHub App to automatically trigger analysis on new PRs across a GitHub organization, and surfaces submitter and reviewer identity in each finding.
    • Connects AI Security Code Review to Endor Labs' policy engine and existing SCMSCMSource Code Management: version-control systems (Git, SVN, etc.) that track changes to code over time, giving cyber tools reproducible builds, audit trails, and collaborative development workflows. and project management systems to automate ticket creation and routing — for example, sending compliance-related PII findings to GRCGRCGovernance, Risk, and Compliance — a framework for aligning IT controls with business policy, regulatory requirements, and risk management so organizations can audit and report on their posture systematically. teams or infrastructure changes to DevOps.

    5 details in Everything

35. Endor Labs + SpaceXAI: Securing every stage of agentic software delivery | Blog | Endor Labs (2026-10-06) Oct 6, 2026 · issue 099

    Endor Labs Agent Kit brings AURI remediation agents (AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. and SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself.) into Grok Build sessions and CI, with per-change approval gates and audit-ready evidence.

    • New integrationAdds Endor Labs Agent Kit for Grok Build, installable via /add-plugin endorlabs in the agent chat or from the marketplace, bringing specialized AURI remediation agents directly into the IDE and CI/cloud runs through the Grok Build SDK.
    • Adds AI SAST Remediation agent (part of the Endor Labs Agent Kit) that traces data flow from untrusted input to vulnerable calls, proposes fixes and test plans, and can open a pull request upon developer approval.
    • New detectionAdds SCA Remediation agent (part of the Endor Labs Agent Kit) that queries scan results, checks package usage, applies upgrades or Endor Patches, adds tests, and rescans to confirm no new vulnerabilities were introduced.
    • New detectionAdds Endor Scores rating system for MCPMCPModel Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. servers and skills, mapped to the OWASPOWASPA nonprofit foundation that publishes open standards, guides, and ranked vulnerability lists—most famously the OWASP Top 10—giving cyber tools a vendor-neutral benchmark to test or report against. Top 10 for LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. Applications, giving security teams a trust signal for every agent component in use.

    4 details in Everything

36. Endor Labs + SpaceXAI: Securing every stage of agentic software delivery | Blog | Endor Labs (2026-10-06) Oct 6, 2026 · issue 016

    Endor Labs Agent Kit brings AURI remediation agents (AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. and SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself.) into Grok Build sessions and CI, with per-change approval gates and audit-ready evidence.

    • New integrationAdds Endor Labs Agent Kit for Grok Build, installable via /add-plugin endorlabs in the agent chat or from the marketplace, bringing specialized AURI remediation agents directly into the IDE and CI/cloud runs through the Grok Build SDK.
    • Adds AI SAST Remediation agent (part of the Endor Labs Agent Kit) that traces data flow from untrusted input to vulnerable calls, proposes fixes and test plans, and can open a pull request upon developer approval.
    • New detectionAdds SCA Remediation agent (part of the Endor Labs Agent Kit) that queries scan results, checks package usage, applies upgrades or Endor Patches, adds tests, and rescans to confirm no new vulnerabilities were introduced.
    • New detectionAdds Endor Scores rating system for MCPMCPModel Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. servers and skills, mapped to the OWASPOWASPA nonprofit foundation that publishes open standards, guides, and ranked vulnerability lists—most famously the OWASP Top 10—giving cyber tools a vendor-neutral benchmark to test or report against. Top 10 for LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. Applications, giving security teams a trust signal for every agent component in use.

    4 details in Everything

37. docs update Oct 6, 2026 · issue 096

    Endor Labs adds Package Firewall user attribution and Claude Code compliance integration for agent governance.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    • Adds Connect the Anthropic Compliance API integration to enroll a Claude Enterprise directory so Coding Agent Governance can match Claude Code sessions to organization roles and groups.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Adds Package Firewall user attribution to record the developer behind each request instead of the shared API key.

    2 details in Everything

38. Endor Labs + SpaceXAI: Securing every stage of agentic software delivery | Blog | Endor Labs (2026-10-05) Oct 5, 2026 · issue 096

    Endor Labs introduces AURI Agent Kit, Coding Agent Governance hooks, and an MCPMCPModel Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. server to secure agentic code delivery on Grok Build end to end.

    • New integrationInstall the Endor Labs Agent Kit into Grok Build via the marketplace or by typing /add-plugin endorlabs in the agent chat to enable in-session security scanning, triage, and remediation.
    • Adds Coding Agent Governance, which uses deterministic hooks to record every MCP tool call, shell command, and file operation and enforce policies before each action runs — without relying on a model's judgment.
    • Adds the Endor Labs Agent Kit, bringing specialized AURI remediation agents (AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. Remediation and SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself. Remediation) directly into Grok Build sessions and CI/cloud runs via the Grok Build SDK.
    • Adds a hooks integration that scans every package an agent installs and blocks confirmed malicious packages at install time.
    • Adds an MCP server that checks for code security risks, exposed secrets, and vulnerable dependencies directly from within an agent session.

    5 details in Everything

    • New detection MCP server and skill trust scoring via Endor Scores mapped to OWASP Top 10 for LLM Applications
39. docs update Oct 2, 2026 · issue 092

    Endor Labs adds beta License Notice reports in TXT and Markdown, exporting license texts, copyright notices, and attribution data for up to 100 projects or 500 packages.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    └──▷ HOW TO FIND IT
    Generate a license attribution file in Markdown before distributing software, scoping to specific projects and filtering to declared-only, permissive or copyleft licenses with a legal disclaimer prepended.
    📍In the Endor Labs console, open Reports and click the dropdown arrow next to the create button, then select License Notice Report. Set output format to Markdown. Under Scope, choose Select Projects and pick up to 100 projects. Set Type to Declared Only, License Category to Permissive and Copyleft, enable Include Copyright information, and enter your legal text in 'Use custom disclaimer to the notice file' (up to 8,192 characters). Click Create.
    • Adds beta License Notice Report type (TXT and Markdown output) to export open-source dependency license texts, copyright notices, and source code locations for distribution compliance.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Supports scoping a License Notice report to Select Projects (up to 100) or Select Packages (up to 500), with a validation error surfaced when either limit is exceeded.
    • Adds Multi-Licenses filter to License Notice reports with All or Most Permissive options to control how dependencies with multiple licenses are represented.
    • Adds Type filter to License Notice reports with All, Declared Only, and Discovered Only options to distinguish between declared and scanning-discovered licenses.
    • Adds License Category filter (e.g. Permissive, Copyleft, Public Domain) to License Notice reports; all categories included by default with a Reset filters option.

    +3 more show less

    • Adds Include Copyright information toggle to License Notice reports, which includes copyright statements per dependency and retains dependencies that have copyright info but no license.
    • Adds Use custom disclaimer to the notice file field (up to 8,192 characters) to prepend custom disclaimer text to a License Notice report.
    • Analytics reports are available in JSON format; Findings reports are available in PDF.
    • Adds beta License Notice Report type (TXT and Markdown output) to export open-source dependency license texts, copyright notices, and source code locations for distribution compliance.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Supports scoping a License Notice report to Select Projects (up to 100) or Select Packages (up to 500), with a validation error surfaced when either limit is exceeded.
    • Adds Multi-Licenses filter to License Notice reports with All or Most Permissive options to control how dependencies with multiple licenses are represented.
    • Adds Type filter to License Notice reports with All, Declared Only, and Discovered Only options to distinguish between declared and scanning-discovered licenses.
    • Adds License Category filter (e.g. Permissive, Copyleft, Public Domain) to License Notice reports; all categories included by default with a Reset filters option.

    +2 more show less

    • Adds Include Copyright information toggle to License Notice reports, which includes copyright statements per dependency and retains dependencies that have copyright info but no license.
    • Adds Use custom disclaimer to the notice file field (up to 8,192 characters) to prepend custom disclaimer text to a License Notice report.

    1 more in Everything

40. docs update Oct 2, 2026 · issue 092

    Endor Labs: Generates license notice reports in TXT or Markdown format, listing open-source dependencies with license texts, copyright notices, and source code locations — required when distributing software containing open-source dependencies.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    └──▷ HOW TO FIND IT
    Generate a Markdown license notice report scoped to copyleft dependencies only, with copyright statements and a custom disclaimer, before shipping a distribution.
    📍In the Endor Labs console, navigate to a project, select Generate License Notice Report, set output format to Markdown, set License Category to Copyleft, enable Include Copyright information, enter disclaimer text in 'Use custom disclaimer to the notice file', then click Generate.
    • Adds license source filter (All, Declared Only, Discovered Only) to control whether declared or scan-discovered licenses appear in the report.
    • Supports scoping a report to one or more specific packages within a project; selecting multiple packages produces a single combined report, and test dependencies are always excluded.
    • Supports generating a single license notice report across multiple projects or packages.

    3 details in Everything

41. docs update Oct 2, 2026 · issue 092

    Endor Labs adds Claude Code session governance via Anthropic Compliance API integration⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    • Integrates with the Anthropic Compliance API to enroll Claude Enterprise directories, enabling Coding Agent Governance to match Claude Code sessions to organization roles and groups.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    1 detail in Everything

42. docs update Sep 21, 2026 · issue 081

    Endor Labs adds Conan package manager support for C/C++ SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself. and NuGetNuGetMicrosoft's package manager for .NET, hosting a public registry of reusable libraries so tools can pull in dependencies without bundling or manually maintaining third-party code. blocking in Package Firewall.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    • Supports SCA for C/C++ projects using the Conan package manager, resolving full transitive dependency graphs from conanfile.txt and conanfile.py manifest files, with vulnerability mapping from OSVOSVOpen Source Vulnerabilities, a vulnerability database and schema maintained by Google that gives cyber tools a machine-readable, cross-ecosystem format for querying and sharing CVE-style advisories., NVDNVDThe National Vulnerability Database, maintained by NIST, is the U.S. government repository of CVE records enriched with CVSS scores and metadata; cyber tools query it to drive vulnerability prioritization., and other sources, including private registries such as JFrog Artifactory and AWS CodeArtifact.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Enables Package Firewall to block malicious NuGet packages during installation, with integration via direct connection, JFrog Artifactory, or Sonatype Nexus Repository.

    2 details in Everything

43. docs update Sep 21, 2026 · issue 081

    Endor Labs adds deploy hooks for GitHub Copilot CLI and VS Code agent mode to capture AI coding session events.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

44. docs update Sep 17, 2026 · issue 077

    Endor Labs adds Threat Center, VS Code extension firewall, malware exposure classification, and background SBOMSBOMA software bill of materials, an inventory of the components and versions inside an application. It helps teams find affected software when a dependency has a vulnerability or licensing issue./VEXVEXA machine-readable statement from a vendor saying whether a CVE in a component actually affects their product. It is how a scanner is told to stop flagging a vulnerability that is not reachable. exports⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    └──▷ HOW TO FIND IT
    Raise findings for dependencies whose version constraints could resolve to a malicious package — catching exposure before the bad version is actually pinned.
    📍In the Endor Labs console, go to Malware finding policy › configure Potentially Affected Finding Severity to your desired severity level (e.g. High), then save the policy.
    • Adds Potentially Affected Finding Severity parameter to the Malware finding policy, letting you raise findings for dependencies whose version constraints could resolve to a malicious package version even when the currently resolved version is clean.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Adds Threat Center, a tenant-wide dashboard tracking newly discovered malware, open-source packages, vulnerabilities, and AI models — including which monitored projects may be exposed before risk surfaces in project-level scans.
    • Extends Package Firewall to block or warn on malicious VS Code extensions from the Microsoft Marketplace, enforcing minimum package age and malware policy before installation; deployable via MDMMDMMobile device management, centrally configuring and managing enrolled devices. Administrators use it to apply settings, distribute applications and enforce device security policies. tools such as Microsoft Intune, Jamf, or Kandji.
    • Classifies malware findings as Definitely Affected or Potentially Affected based on how precisely a dependency matches a known malicious package version.
    • Coding Agent Governance policies with the Ask Permission action in Codex now let the tool call proceed and record the match under Policy Violations as Ask, rather than denying the call outright; Block policies continue to deny.

    +1 more show less

    • Generates SBOM and VEX exports as background jobs so large projects can export without the request timing out; exports are available per project, finding, or package version.
    • Adds Potentially Affected Finding Severity parameter to the Malware finding policy, letting you raise findings for dependencies whose version constraints could resolve to a malicious package version even when the currently resolved version is clean.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Adds Threat Center, a tenant-wide dashboard tracking newly discovered malware, open-source packages, vulnerabilities, and AI models — including which monitored projects may be exposed before risk surfaces in project-level scans.
    • Classifies malware findings as Definitely Affected or Potentially Affected based on how precisely a dependency matches a known malicious package version.

    3 more in Everything

    └──▷ BREAKING ON UPGRADE

    • !Coding Agent Governance Ask Permission policies in Codex no longer block tool calls — calls now proceed instead of being denied. Any policy you relied on to stop calls in Codex must be changed to Block.
45. docs update Sep 17, 2026 · issue 077

    Endor Labs adds Coding Agent Governance with MDMMDMMobile device management, centrally configuring and managing enrolled devices. Administrators use it to apply settings, distribute applications and enforce device security policies. deployment and Claude Code compliance API support⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    • Adds Deploy Coding Agent Governance with MDM support, enabling generation of Cursor hooks, Claude Code settings, Codex requirements, GitHub Copilot policy hooks, or a macOS configuration profile for managed developer machines.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Adds Connect the Anthropic Compliance API integration to enroll a Claude Enterprise directory so Coding Agent Governance can match Claude Code sessions to organization roles and groups.
    • Adds Understand blocked package responses documentation covering each Package Firewall block reason, its meaning, and resolution steps.
    • Adds Deploy Coding Agent Governance with MDM support, enabling generation of Cursor hooks, Claude Code settings, Codex requirements, GitHub Copilot policy hooks, or a macOS configuration profile for managed developer machines.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Adds Connect the Anthropic Compliance API integration to enroll a Claude Enterprise directory so Coding Agent Governance can match Claude Code sessions to organization roles and groups.

    1 more in Everything

46. Threat Center and Package Firewall for VS Code Extensions | Blog | Endor Labs (2026-09-16) Sep 16, 2026 · issue 077

    Endor Labs launches Threat Center for real-time malware monitoring and extends Package Firewall to VS Code extensions.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    • Adds Threat Center, a real-time view of malicious packages landing on public registries (npm, PyPI, and others), with impact mapping showing which of your projects pull in an affected package, malware behavior details, and alerts regardless of whether you are impacted.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Extends Package Firewall to VS Code extensions, scanning every new extension or update against the policy engine and blocking disallowed extensions before they appear in the marketplace inside the editor.
    • Adds research context to Threat Center entries, with the security team supplementing automated analysis with deeper write-ups, blog links, and ongoing updates as a campaign develops.
    • Adds Threat Center, a real-time view of malicious packages landing on public registries (npm, PyPI, and others), with impact mapping showing which of your projects pull in an affected package, malware behavior details, and alerts regardless of whether you are impacted.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    2 more in Everything

47. docs update Sep 14, 2026 · issue 074

    Endor Labs adds Potentially Affected malware exposure classification and a severity threshold policy parameter for unpinned dependencies.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    └──▷ HOW TO FIND IT
    Tune the malware policy to raise findings when an unpinned dependency could resolve to a malicious version — useful for catching dev-vs-CI environment drift before it becomes an incident.
    📍In the Endor Labs console, navigate to the Malware finding policy › Parameters, set Potentially Affected Finding Severity to High, then save. Enable Upgrade Policies to Latest Version under Configure policy settings to apply this to existing tenants automatically.
    • Adds a Potentially Affected Finding Severity parameter to the Malware finding policy, configurable to Critical, High, Medium, Low, or None, to control whether findings are raised for unpinned dependencies whose declared version range could resolve to a known malicious package version.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
    • Adds an Exposure parameter to finding policy templates to include or exclude findings based on whether a dependency is Definitely Affected or Potentially Affected.
    • Introduces a Potentially Affected malware exposure classification for unpinned dependencies whose resolved version does not currently match a malicious package but whose declared version range could resolve to one across environments or future resolution runs.

    3 details in Everything

48. docs update Sep 12, 2026 · issue 072

    Endor Labs adds base image remediation with upgrade comparison, Poetry private registry auth, and async SBOMSBOMA software bill of materials, an inventory of the components and versions inside an application. It helps teams find affected software when a dependency has a vulnerability or licensing issue./VEXVEXA machine-readable statement from a vendor saying whether a CVE in a component actually affects their product. It is how a scanner is told to stop flagging a vulnerability that is not reachable. export via AsyncJob API.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

    └──▷ TRY IT
    Identify which base image upgrade resolves the most container findings before committing to an update in CI.
    copy

    $ endorctl container scan --base-image-check-updates && endorctl container remediation list-base-image-updates

    • Adds endorctl container scan --base-image-check-updates to attribute container findings to base vs.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ application layers and compare remediation options across a refreshed build, next version, and latest version.
    • Adds endorctl container remediation list-base-image-updates to list base image upgrade options with per-option counts of findings fixed and introduced.
    • Adds AsyncJob REST APIREST APIA web API design style where clients interact with resources over HTTP using standard methods like GET, POST, and DELETE, giving cyber tools a predictable, language-agnostic way to integrate with external services. resource for submitting and polling SBOM and VEX export jobs as background tasks, enabling large-project exports without request timeouts.
    • Passes configured PyPI package manager integration credentials to Poetry during scans when a [[tool.poetry.source]] URL matches an integration, authenticating poetry lock and poetry install against private registries.
    • Adds endorctl container scan --base-image-check-updates to attribute container findings to base vs.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ application layers and compare remediation options across a refreshed build, next version, and latest version.
    • Adds endorctl container remediation list-base-image-updates to list base image upgrade options with per-option counts of findings fixed and introduced.
    • Adds AsyncJob REST APIREST APIA web API design style where clients interact with resources over HTTP using standard methods like GET, POST, and DELETE, giving cyber tools a predictable, language-agnostic way to integrate with external services. resource for submitting and polling SBOM and VEX export jobs as background tasks, enabling large-project exports without request timeouts.

    1 more in Everything

49. docs update Sep 7, 2026 · issue 067

    Endor Labs adds Package Firewall for VS Code extensions and incremental container scanning to focus on build-introduced findings.

    • Extends Package Firewall to block or warn on malicious VS Code extensions from the Microsoft Marketplace before they install, deployable via MDMMDMMobile device management, centrally configuring and managing enrolled devices. Administrators use it to apply settings, distribute applications and enforce device security policies. tools such as Microsoft Intune, Jamf, or Kandji.
    • Adds incremental container scanning to compare a new build against a baseline scan and surface only findings the build introduced, fixed, or shares with the baseline — with the option to suppress shared findings so only new ones remain.
    • Adds incremental container scanning to compare a new build against a baseline scan and surface only findings the build introduced, fixed, or shares with the baseline — with the option to suppress shared findings so only new ones remain.

    1 more in Everything

50. docs update Sep 7, 2026 · issue 067

    Endor Labs Package Firewall now blocks malicious VS Code extension installs from the Microsoft Marketplace via MDMMDMMobile device management, centrally configuring and managing enrolled devices. Administrators use it to apply settings, distribute applications and enforce device security policies. deployment.

51. How Endor Labs AI SAST Learns What Matters to Your Organization | Blog | Endor Labs (2026-09-07) Sep 7, 2026 · issue 067

    Endor Labs AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. now ingests IaCIaCInfrastructure as Code: the practice of defining and provisioning compute infrastructure through machine-readable config files rather than manual processes, giving cyber tools repeatable, version-controlled environment setup., repo context files, user feedback, and uploaded threat models to continuously sharpen finding prioritization.

    • Automatically parses AGENTS.md, CLAUDE.md, skills files, and AI rules files in the repository to extract security-relevant context (authentication model, internal vs. external users) for every scan.
    • Automatically ingests Dockerfiles, Terraform, CloudFormationCloudFormationAn AWS service that lets you describe and provision cloud infrastructure as declarative YAML or JSON templates, so a cyber tool can reproducibly spin up or tear down its entire AWS environment with a single API call., Kubernetes YAML, and Helm charts to build a deployment model used during PR scans and local developer scans, with no configuration required.
    • Enables users to upload architecture and threat model context as plain natural language, which AI SAST verifies against the codebase and auto-classifies into project context, true-positive patterns, and false-positive patterns, scopable to a single project, a set of projects, or the entire tenant.
    • Adds per-finding thumbs-up / thumbs-down feedback during triage so teams can teach AI SAST which issue classes matter most, adjusting severity scoring for entire categories in future scans.

    4 details in Everything

52. docs update Sep 4, 2026 · issue 064

    Endor Labs extends Bazel rules_apple SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself. to watchOS, tvOS, and visionOS targets.

53. Introducing Package Firewall | Blog | Endor Labs (2026-09-03) Sep 3, 2026 · issue 064

    Endor Labs Package Firewall blocks malicious, vulnerable, and non-compliant packages before install across npm, PyPI, Go, and more.

    • Adds Package Firewall, an enforcement layer that intercepts every package install request and returns a 403 for flagged packages — the package never touches disk — while clean packages are redirected to the upstream registry normally.
    • Supports integration with private artifact managers (e.g., JFrog Artifactory) by pointing a remote repository at the Package Firewall URL, requiring no changes on developer machines.
    • Supports Endpoint Mode, where Package Firewall configuration is pushed to developer machines via MDMMDMMobile device management, centrally configuring and managing enrolled devices. Administrators use it to apply settings, distribute applications and enforce device security policies. and to CI runners via a config snippet, requiring no manual action from developers.
    • Enforces cooldown periods, blocking newly published package versions for a configurable window regardless of requester, addressable at scale via policy.
    • Integrates with the Endor Labs policy engine to block packages based on known vulnerabilities (CVEs) or license non-compliance, beyond malware alone.

    +2 more show less

    • Logs every install event — package name, version, time, and reason — visible in the Endor Labs UI and via the API.
    • Powered by a real-time malware feed with AI-enhanced multi-pass analysis (LLMsLLMsA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code., static analysis, behavioral signals) covering npm, PyPI, and Go, classifying packages within minutes of upload.

    7 details in Everything

54. announced Sep 3, 2026 · issue 064

    Endor Labs: New Package Firewall capability intercepts every package install request and returns a 403 to block flagged packages before they touch disk, with clean packages redirected transparently to the upstream registry.

    • Enforces the full Endor Labs policy engine at install time: blocks packages with known CVEs or packages that violate license compliance policies.
    • Powered by a real-time malware feed that continuously scans every newly uploaded package across npm, PyPI, Go, and other ecosystems using multi-pass LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. and static analysis, classifying packages within minutes of upload.

    2 details in Everything

55. announced Sep 3, 2026 · issue 010

    Endor Labs: New Package Firewall capability intercepts every package install request and returns a 403 to block flagged packages before they touch disk, with clean packages redirected transparently to the upstream registry.

    • Enforces the full Endor Labs policy engine at install time: blocks packages with known CVEs or packages that violate license compliance policies.
    • Powered by a real-time malware feed that continuously scans every newly uploaded package across npm, PyPI, Go, and other ecosystems using multi-pass LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. and static analysis, classifying packages within minutes of upload.

    2 details in Everything

56. Introducing Package Firewall | Blog | Endor Labs (2026-09-03) Sep 3, 2026 · issue 010

    Endor Labs Package Firewall blocks malicious, vulnerable, and non-compliant packages before install across npm, PyPI, Go, and more.

    • Adds Package Firewall, an enforcement layer that intercepts every package install request and returns a 403 for flagged packages — the package never touches disk — while clean packages are redirected to the upstream registry normally.
    • Supports integration with private artifact managers (e.g., JFrog Artifactory) by pointing a remote repository at the Package Firewall URL, requiring no changes on developer machines.
    • Supports Endpoint Mode, where Package Firewall configuration is pushed to developer machines via MDMMDMMobile device management, centrally configuring and managing enrolled devices. Administrators use it to apply settings, distribute applications and enforce device security policies. and to CI runners via a config snippet, requiring no manual action from developers.
    • Enforces cooldown periods, blocking newly published package versions for a configurable window regardless of requester, addressable at scale via policy.
    • Integrates with the Endor Labs policy engine to block packages based on known vulnerabilities (CVEs) or license non-compliance, beyond malware alone.

    +2 more show less

    • Logs every install event — package name, version, time, and reason — visible in the Endor Labs UI and via the API.
    • Powered by a real-time malware feed with AI-enhanced multi-pass analysis (LLMsLLMsA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code., static analysis, behavioral signals) covering npm, PyPI, and Go, classifying packages within minutes of upload.

    7 details in Everything

57. docs update Sep 1, 2026 · issue 061

    Endor Labs now surfaces native package-manager dependency scopes (e.g. Maven provided, npm devDependencies) on findings and metadata.

    • Adds the un-normalized, package-manager-native dependency scope (e.g. Maven provided, npm devDependencies) to findings and dependency metadata after a scan, filterable via the API and usable in exception policies.

    1 detail in Everything

58. docs update Aug 31, 2026 · issue 060

    Endor Labs adds a real-time proxy that intercepts package manager requests and blocks known malicious packages before installation.

    • Adds a real-time malicious-package interception layer that sits between package manager clients and public registries, blocking known malicious packages at install time while passing safe packages through unchanged.

    1 detail in Everything

59. docs update Aug 31, 2026 · issue 060

    Endor Labs adds Bazel dependency resolution for iOS/macOS apps built with rules_apple 4.0.0+

    • Supports SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself. scanning of Bazel ios_application, ios_framework, macos_application, and related bundle and test rules built with rules_apple 4.0.0 or later, covering Swift, Objective-C, and prebuilt framework/XCFramework dependencies; scans must run on a macOS host with Bazel aspects enabled.

    1 detail in Everything

60. Why Endor Labs AI SAST for C finds what other tools miss | Blog | Endor Labs (2026-08-24) Aug 24, 2026 · issue 056

    Endor Labs AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. adds full C language support with buildless scanning in the IDE and on PRs.

    • Adds full C language support to Endor Labs AI SAST, enabling buildless scans that run in the IDE and on the PR without requiring a compiler hook or reproducible build.
    • Pairs deterministic call-graph and reachability program analysis with LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. reasoning to trace data flow across functions and files, catching buffer overflows, use-after-free, and integer overflows that span multiple functions.
    • Each C finding includes a call path, a working exploit, and a suggested fix — consistent with the evidence format for all other supported languages.
    • C SAST and C SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself. run together in the same platform, covering first-party C code and its open-source dependencies side by side.
    • C scanning runs as part of AURI by Endor Labs, providing an independent security harness outside the coding agent with policy enforcement across any agent, model, or CI stage.

    5 details in Everything

61. Why Endor Labs AI SAST for C finds what other tools miss | Blog | Endor Labs (2026-08-19) Aug 19, 2026 · issue 050

    Endor Labs AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. adds full C language support with buildless scanning in the IDE and on PRs.

    • Adds full C language support to Endor Labs AI SAST, with buildless scans that run in the IDE and on pull requests — no compiler hook or reproducible build required.
    • Pairs deterministic program analysis (call-graph and reachability engine) with LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. reasoning for C, tracing data flow across functions and files to catch buffer overflows, use-after-free, integer overflows, and other cross-function memory-safety flaws that pattern-based rules miss.
    • C SAST is available alongside C SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself. in the same platform, covering first-party C code and its open-source dependencies together.
    • Surfaces C scanning through AURI by Endor Labs, providing an independent security harness outside the coding agent with policy enforced across any agent, model, or CI stage.

    4 details in Everything

62. Why Endor Labs AI SAST for C finds what other tools miss | Blog | Endor Labs (2026-08-19) Aug 19, 2026 · issue 008

    Endor Labs AI SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. adds full C language support with buildless scanning in the IDE and on PRs.

    • Adds full C language support to Endor Labs AI SAST, with buildless scans that run in the IDE and on pull requests — no compiler hook or reproducible build required.
    • Pairs deterministic program analysis (call-graph and reachability engine) with LLMLLMA large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. reasoning for C, tracing data flow across functions and files to catch buffer overflows, use-after-free, integer overflows, and other cross-function memory-safety flaws that pattern-based rules miss.
    • C SAST is available alongside C SCASCASoftware composition analysis, identifying third-party dependencies and checking them for known vulnerabilities and license issues. It covers software a team uses rather than writes itself. in the same platform, covering first-party C code and its open-source dependencies together.
    • Surfaces C scanning through AURI by Endor Labs, providing an independent security harness outside the coding agent with policy enforced across any agent, model, or CI stage.

    4 details in Everything

63. docs update Aug 19, 2026 · issue 050

    Endor Labs adds Agents Hub for AI agent discovery and Agent Governance for monitoring and policy enforcement over coding agents.

    • Adds Agents Hub — a catalog where practitioners can browse, filter by category, and select published Endor Labs security agents (covering workflows such as finding triage, dependency remediation, and scan-failure diagnosis), and monitor team usage including task counts and last-run time.
    • Adds Agent Governance — an inventory and control plane for AI coding agents, models, MCPMCPModel Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. servers, and skills running across developer workstations and cloud environments, with real-time monitoring of agent actions (commands, file access, tool calls).
    • Adds policy enforcement in Agent Governance that can block, alert on, or require approval for risky agent behavior before it executes, with a built-in policy catalog extensible by custom rules.
    • Adds reusable filters for findings and projects under Coding Agent Governance, letting users save and apply custom views and criteria across their data.

    4 details in Everything

64. When you can't upgrade: open source examples of Endor Patches | Blog | Endor Labs (2026-08-18) Aug 18, 2026 · issue 050

    Endor Labs open-sources two complete Endor Patch examples with reproducible builds, attestations, and auditable diffs in the endor-patches repository.

    └──▷ TRY IT
    Independently verify a patch's security diffs by inspecting the attestation file and matching upstream commit SHAs before installing.
    copy

    $ cat patches/com.fasterxml.woodstox/woodstox-core/security_attestation.json | jq '.[] | select(.fix_type == "FIX_TYPE_VULN")'

    Reproduce the patched artifact from scratch and verify its checksum matches the published artifact.
    copy

    $ # Extract the source tarball, then build with Docker and Bazel per the included README.md
    tar -xf patches/com.fasterxml.woodstox/woodstox-core/source.tar.gz
    cd woodstox-core-patched && docker run --rm -v "$PWD":/build <hermetic-image> bazel build //...
    sha256sum output/woodstox-core-6.3.1-endor-2024-09-03.jar
    diff <(sha256sum output/woodstox-core-6.3.1-endor-2024-09-03.jar | awk '{print $1}') SHA256SUM.txt

    Install the patched JAR into your local Maven repository and override a transitive dependency so nothing else in the build needs to change.
    copy

    $ mvn install:install-file \
      -Dfile=patches/com.fasterxml.woodstox/woodstox-core/artifacts/woodstox-core-6.3.1-endor-2024-09-03.jar \
      -DgroupId=com.fasterxml.woodstox \
      -DartifactId=woodstox-core \
      -Dversion=6.3.1-endor-2024-09-03 \
      -Dpackaging=jar

    • Patched JARs in patches/<library>/artifacts/ can be installed to a local Maven repository via mvn install:install-file and declared as dependencies using the -endor-YYYY-MM-DD version string, with transitive override support via <dependencyManagement> in Maven or resolutionStrategy in Gradle.

    1 detail in Everything

65. When you can't upgrade: open source examples of Endor Patches | Blog | Endor Labs (2026-08-18) Aug 18, 2026 · issue 008

    Endor Labs open-sources two complete Endor Patch examples with reproducible builds, attestations, and auditable diffs in the endor-patches repository.

    └──▷ TRY IT
    Independently verify a patch's security diffs by inspecting the attestation file and matching upstream commit SHAs before installing.
    copy

    $ cat patches/com.fasterxml.woodstox/woodstox-core/security_attestation.json | jq '.[] | select(.fix_type == "FIX_TYPE_VULN")'

    Reproduce the patched artifact from scratch and verify its checksum matches the published artifact.
    copy

    $ # Extract the source tarball, then build with Docker and Bazel per the included README.md
    tar -xf patches/com.fasterxml.woodstox/woodstox-core/source.tar.gz
    cd woodstox-core-patched && docker run --rm -v "$PWD":/build <hermetic-image> bazel build //...
    sha256sum output/woodstox-core-6.3.1-endor-2024-09-03.jar
    diff <(sha256sum output/woodstox-core-6.3.1-endor-2024-09-03.jar | awk '{print $1}') SHA256SUM.txt

    Install the patched JAR into your local Maven repository and override a transitive dependency so nothing else in the build needs to change.
    copy

    $ mvn install:install-file \
      -Dfile=patches/com.fasterxml.woodstox/woodstox-core/artifacts/woodstox-core-6.3.1-endor-2024-09-03.jar \
      -DgroupId=com.fasterxml.woodstox \
      -DartifactId=woodstox-core \
      -Dversion=6.3.1-endor-2024-09-03 \
      -Dpackaging=jar

    • Patched JARs in patches/<library>/artifacts/ can be installed to a local Maven repository via mvn install:install-file and declared as dependencies using the -endor-YYYY-MM-DD version string, with transitive override support via <dependencyManagement> in Maven or resolutionStrategy in Gradle.

    1 detail in Everything