https://cybertoolchain.io/tools/jfrog New releases and features in JFrog, tracked by The Cyber Toolchain. en Wed, 21 Oct 2026 06:25:58 GMT https://jfrog.com/blog/shadow-ai-detection-extends-to-developer-workstations/ https://jfrog.com/blog/shadow-ai-detection-extends-to-developer-workstations/ Wed, 21 Oct 2026 06:25:58 GMT JFrog AI Security extends Shadow AI detection to developer workstations via the JFrog Agent Plugin.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ • Adds the `JFrog Agent Plugin` installable on developer workstations to discover and locally scan AI assets — skills, MCP servers, plugins, hooks, commands, and monitors — that bypass central repository controls.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ • Introduces a three-step local detection pipeline: Discover (enumerate AI assets on the workstation), Scan Locally (deterministic rules-based scan), and Report (sends metadata, findings, and risk analysis to the JFrog Platform). • Adds a unified AI asset inventory in JFrog AI Security showing each asset's name, type, reporting harness (Claude Code, Cursor, etc.), workstation count, local risk analysis, JFrog origin, and detection timestamp. • Adds risk assessment classifications for detected assets: Shell Access + WebFetch (high exfiltration risk), Unscoped Write/Edit (writes outside designated directories), and BypassPermissions Mode (executes without required approval). • Adds an Origin field per asset indicating whether it matches an artifact already managed in JFrog Artifactory or JFrog Catalog, or has 'No Match' against the JFrog Platform. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/jfrog https://jfrog.com/blog/scanning-ai-assets/ https://jfrog.com/blog/scanning-ai-assets/ Wed, 21 Oct 2026 06:18:40 GMT JFrog adds intent-based AI asset scanning for SKILL.md files, agent plugins, and APM packages via JFrog AI Security and Agent Guard.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ • Adds `JFrog AI Security` scanner that performs semantic (intent-based) analysis of agent skill files (e.g.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ `SKILL.md`), agent plugins, and APM manifest packages to detect prompt injection, hidden instructions, obfuscated payloads, and credential-exfiltration instructions that SCA and SAST cannot reach. • Adds `JFrog Agent Guard` enforcement component that applies deterministic-plus-semantic policy at the agent harness — the point an agent actually invokes the asset — not only at download time, so a skill reclassified after initial approval is blocked on machines that already have it. • Returns one of three scan classifications — 'Not Malicious', 'Suspicious', or 'Malicious' — for each AI asset, with policy controls that route outcomes to block, review queue, or allow. • Supports per-version allow-lists so approving one version of a skill does not automatically approve subsequent versions. • Supports enforcement by severity threshold (e.g. block any skill with a Critical-severity finding) and time-bound waivers with fixed expiration dates. • Stores agent skills, plugins, and APM packages in `JFrog Artifactory` alongside Docker images and npm packages under the same permissions model and audit trail, unifying AI and software supply chain assets in a single platform. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/jfrog https://jfrog.com/blog/jfrog-business-value-center/ https://jfrog.com/blog/jfrog-business-value-center/ Thu, 15 Oct 2026 13:15:56 GMT JFrog Business Value Center adds a real-time ROI and supply chain coverage dashboard inside MyJFrog for SaaS customers.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ • Adds the JFrog Business Value Center inside `MyJFrog` — a real-time dashboard that quantifies platform ROI, developer hours reclaimed, storage saved via deduplication, and the percentage of repositories actively scanned by JFrog Xray.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ • Surfaces an Optimization panel showing what share of the software supply chain is scanned for vulnerabilities and what fraction of incoming open-source packages are blocked by JFrog Curation policies before reaching developers. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/jfrog https://jfrog.com/blog/governing-the-agentic-sdlc/ https://jfrog.com/blog/governing-the-agentic-sdlc/ Mon, 12 Oct 2026 16:24:04 GMT JFrog introduces the Agentic Process Evidence Protocol and AppTrust Prompt-to-Release Traceability to govern AI agent activity across the SDLC.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ • Adds `Prompt to Release Traceability` in JFrog AppTrust, connecting AI agent session logs to signed SLSA-compatible provenance attestations and binding them to git commits, artifacts, and released versions.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ • Introduces the Agentic Process Evidence Protocol, a four-step framework that captures agent session telemetry (user prompts, thinking tokens, file edits, tool calls), cryptographically signs and attaches evidence to SDLC entities, evaluates evidence at policy gates before promotion, and persists session logs for post-release auditing. • Enables policy-as-code release gates that evaluate Agentic Process Evidence to enforce checks such as approved LLM usage, approved MCP tool usage, human accountability, human review presence, and absence of unapproved lateral operations. • Captures five signed categories of facts per agentic process: agent identity (agent version, harness version, model ID), execution traceability (full session logs as structured JSON covering every tool call, API call, input, and context source), provenance, integrity, and policy compliance. • Publishes an open `Agentic Process Evidence` specification on GitHub (`https://github.com/jfrog/agentic-process-evidence`) for integration by agent frameworks (Cursor, OpenCode, Claude Code, or custom AI workers), regulatory bodies, and SDLC platform providers. • Supports post-release auditing by binding session logs to git commits and software releases, enabling retroactive searches — for example, identifying all releases produced using a later-identified malicious MCP tool. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/jfrog https://jfrog.com/blog/jfrog-artifactory-luarocks-support-nginx-openresty-kong/ https://jfrog.com/blog/jfrog-artifactory-luarocks-support-nginx-openresty-kong/ Wed, 16 Sep 2026 13:15:32 GMT JFrog Artifactory adds native LuaRocks repository support for hosting, proxying, and caching Lua rocks for NGINX, OpenResty, and Kong.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ • Adds LuaRocks as a native repository type in Artifactory, supporting local, remote, and virtual repository tiers for Lua packages (rocks and rockspecs).⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠ • Adds a remote repository type that proxies and caches luarocks.org rocks on first request, so subsequent pulls are served from Artifactory rather than the public registry. • Adds a local repository type for storing private first-party rocks, internal libraries, and custom Kong plugins as versioned `.rock` files. • Adds a virtual repository that unifies local and remote LuaRocks repositories behind a single endpoint for the `luarocks` client. • Enables the `luarocks` CLI to point at an Artifactory virtual repository URL as its registry, replacing direct luarocks.org access for OpenResty and Kong fleets. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/jfrog https://jfrog.com/blog/swampup-2026-recap-single-source-of-truth-agentic-software-supply-chain/ https://jfrog.com/blog/swampup-2026-recap-single-source-of-truth-agentic-software-supply-chain/ Wed, 02 Sep 2026 14:52:19 GMT JFrog extends platform governance to agentic supply chains with new APM Registry, Plugins Registry, Skills Scanning, Agent Guard, and Zero-Touch Remediation. • New APM Repository provides a governed registry for agent package installs, replacing open-internet pulls via `apm install`. • New Plugins Registry delivers a scanned, auditable source for first- and third-party coding agent plugins, replacing default public marketplaces. • New Skills Scanning uses semantic security scanning to detect malicious behavior in skill files and block them before they reach a developer's machine. • New Agent Guard (AI Asset Enforcement) ensures only AI assets approved in AI Catalog are consumed by agents across coding environments. • New Agent Package Resolution provides a trusted, policy-enforced path for agents to resolve dependencies through the JFrog Platform. • New JFrog Agent Plugin integrates coding agents — Claude Code, Cursor, VS Code Copilot, Kiro, Devin, OpenCode, and Codex — natively with the JFrog Platform. • New Traffic Controller integrations with Netskope, Zscaler, and Cloudflare redirect public-repo package requests through JFrog at the network layer. • New JFrog Zero-Touch Remediation automatically evaluates all available fixes for a vulnerability in a customer's environment and selects the best one based on policy, environment, and risk tolerance. • New AutoPR with Smart Remediation automatically pushes verified, context-aware fix guidance directly to pull requests. • Expanded JFrog AI Catalog now governs skills, plugins, and Agent Packages (APM) in addition to previously supported AI assets. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/jfrog https://jfrog.com/blog/jfrog-swampup-2026-live-updates/ https://jfrog.com/blog/jfrog-swampup-2026-live-updates/ Wed, 02 Sep 2026 12:15:14 GMT JFrog swampUP 2026 unveils Zero-Touch Remediation, AI-powered policy-as-code, prompt-to-release traceability, and post-release governance. • Introduces Zero-Touch Remediation via Compliant Version Replacement, which automatically swaps a vulnerable package version for a patched one mid-pipeline by evaluating multiple upstream sources (Chainguard, Maven Central, and others) against the customer's own policy, with JFrog Artifactory as the system of record. • Introduces Agentic Remediation for first-party code, fixing vulnerabilities at the source level with a developer reviewing and approving before merge — distinct from Zero-Touch Remediation's fully automated mid-pipeline swap. • Adds an AI-Powered Policy as Code Playground that converts plain-language intent into enforceable, versioned, Rego-based policy automatically, requiring no manual policy authoring. • Adds Prompt to Release Traceability, shifting evidence collection left to capture the full provenance trail for every agent interaction and artifact — sessions, tickets, pull requests, approvals, and promotions — stored alongside the exact software version being created. • Adds Out-of-the-Box Compliance Frameworks, enabling teams to activate pre-built policy sets for standards such as NIST and CRA and select only the applicable sections, with no custom configuration required. • Adds Repo-Level Policies that enforce evidence-based governance directly on repositories without requiring project, application, or lifecycle-stage setup. • Adds Post-Release Governance, extending compliance monitoring and vulnerability remediation tracking through the support window after a release ships to address European regulatory requirements. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/jfrog https://jfrog.com/blog/prompt-to-release-traceability/ https://jfrog.com/blog/prompt-to-release-traceability/ Wed, 02 Sep 2026 11:56:45 GMT JFrog AppTrust gains Prompt-to-Release Traceability: auto-assembled, signed audit evidence chains for every release. • Launches 'Prompt to Release Traceability' inside JFrog AppTrust, automatically collecting commits, PRs, Jira ticket transitions, and approver records, signing them, and attaching them to the application version before any promotion gate runs. • Validates full Jira ticket transition history (creation → in-progress → done, with dates) rather than merely checking ticket existence, producing auditor-ready workflow proof. • Detects self-approval violations by mapping each PR's merger and approver back to the committer — automatically flagging the pattern most compliance frameworks explicitly prohibit. • Introduces incoming JFrog Agent Plugins to capture AI agent session evidence (prompts, code produced, and a full session bill of materials — including which MCPs, skills, and agentic assets were used) as signed evidence stored in Artifactory. • Launches out-of-the-box compliance frameworks with pre-mapped controls for NIST SSDF and the EU Cyber Resilience Act, ready to enforce without manual configuration. • Git-based evidence (commits, PRs, Jira transitions, and approver chains) is generally available now; agent session evidence and session bill of materials are planned for later this year. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/jfrog https://jfrog.com/blog/the-evolution-of-jfrog-ai-catalog/ https://jfrog.com/blog/the-evolution-of-jfrog-ai-catalog/ Wed, 02 Sep 2026 11:56:37 GMT JFrog AI Catalog expands into a full AI control plane with MCP, skills, plugins, and runtime enforcement for agentic workflows. • Adds an MCP Registry to define which MCP servers are allowed per project, restrict specific tool calls, and scan local MCP packages before an agent can use them. • Adds an Agent Skills Registry to curate, sign, verify, and scan packaged agent behaviors before execution is permitted. • Adds an Agent Plugins Registry to control, sign, verify, and scan plugin bundles agents install. • Adds Agent Packages support to ship an agent's entire context and dependency tree as a single versioned, resolvable package compliant with the open-source APM (Agent Package Manager) standard. • Introduces semantic scans that analyze what an asset instructs an agent to do, rather than matching against known-bad hashes, to catch novel malicious skills and MCP servers. • Adds Shadow AI Detection to automatically detect and categorize every AI asset across an environment as managed, unmanaged, partially managed, or malicious. • Introduces Agent Guard, installed via the official JFrog agent plugin, which enforces organizational policy at runtime by checking every MCP tool call and skill call against defined policies, regardless of agent harness or coding environment. • Releases official JFrog agent plugins for Claude Code, Cursor, VS Code, OpenCode, and Codex that bring organizational policy control directly into the coding agent environment. • Stores all AI asset types — models, MCP servers, skills, plugins, and agent packages — as versioned, traceable artifacts in Artifactory under a single RBAC model and audit trail. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/jfrog https://jfrog.com/blog/jfrog-apptrust-enforces-compliance/ https://jfrog.com/blog/jfrog-apptrust-enforces-compliance/ Wed, 02 Sep 2026 11:55:48 GMT JFrog AppTrust ships Out-of-the-Box Compliance Frameworks for NIST SSDF and EU CRA, with pre-mapped Policy-as-Code enforcement at release gates. • Adds Out-of-the-Box Compliance Frameworks in JFrog AppTrust, launching with NIST SSDF and EU CRA pre-mapped to Policy-as-Code rules — no Rego to write and no manual control mapping required. • Introduces a compliance catalog inside AppTrust where AppSec teams select a framework and see every control translated into plain requirements, each linked to specific Policy-as-Code rules enforced at release gates. • Surfaces coverage gaps explicitly: controls not yet covered by AppTrust are flagged, while controls already satisfied by running AppTrust are shown as covered by default. • Adds full SDLC traceability capturing agent intent sessions, Git commits, pull requests, and Jira tickets, all connected to shipped artifacts, giving Policy-as-Code rules live data to check against. • Separates framework configuration from rollout control: AppSec owns which controls and rules are active; development managers control which applications are in scope, whether rules run as warnings or hard failures, and the warning-to-block conversion window. • Out-of-the-Box Compliance Frameworks are available now as part of the Ultimate Security Bundle. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/jfrog https://jfrog.com/blog/introducing-jfrog-preview/ https://jfrog.com/blog/introducing-jfrog-preview/ Fri, 28 Aug 2026 13:12:14 GMT JFrog launches a Preview early-access track so customers can opt into new capabilities before GA through MyJFrog. • New 'Preview' early-access track delivers two tiers — Closed Preview (invite-only feedback cohorts) and Open Preview (any eligible customer, fully supported, free) — so capabilities reach users as soon as they are production-ready. • Opt-in is self-serve: navigate to `Settings > Preview` in the MyJFrog portal, enable the `Enable Preview` toggle, accept the Preview Terms of Service, and activate individual preview features on your subscription. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/jfrog https://jfrog.com/blog/introducing-agent-package-resolution/ https://jfrog.com/blog/introducing-agent-package-resolution/ Tue, 25 Aug 2026 12:36:35 GMT JFrog Agent Package Resolution (preview) routes AI coding agent package downloads through Artifactory governance automatically. • Introduces Agent Package Resolution (preview), a feature of the JFrog Agent Plugin that routes AI coding agent dependency downloads through JFrog Artifactory instead of public registries, covering npm, PyPI, Maven, Go, Docker Hub, Helm, and NuGet. • Adds session steering: at the start of each Claude Code or Cursor agent session, the plugin automatically supplies the Artifactory URL for each governed package type — no per-project configuration required. • Adds a `jf setup` command for persistent package-manager configuration that binds the local package manager to an Artifactory repository for a given project, persisting across sessions and tools. • Extends JFrog Curation policy enforcement server-side to all agent-driven downloads, blocking malicious packages, unapproved licenses, and policy-violating packages regardless of whether session steering or `jf setup` is active. • Covers indirect and transitive installs — postinstall scripts, `pip install -r requirements.txt`, `npm ci`, and equivalent bulk-pull commands — routing all through Artifactory from the first session. • Produces complete audit trails for every agent-driven download, block, and policy decision alongside existing human-driven build logs. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/jfrog https://jfrog.com/blog/introducing-agent-package-resolution/ https://jfrog.com/blog/introducing-agent-package-resolution/ Tue, 25 Aug 2026 12:36:35 GMT JFrog Agent Package Resolution (preview) routes AI coding agent package downloads through Artifactory governance automatically. • Introduces Agent Package Resolution (preview), a feature of the JFrog Agent Plugin that routes AI coding agent dependency downloads through JFrog Artifactory instead of public registries, covering npm, PyPI, Maven, Go, Docker Hub, Helm, and NuGet. • Adds session steering: at the start of each Claude Code or Cursor agent session, the plugin automatically supplies the Artifactory URL for each governed package type — no per-project configuration required. • Adds a `jf setup` command for persistent package-manager configuration that binds the local package manager to an Artifactory repository for a given project, persisting across sessions and tools. • Extends JFrog Curation policy enforcement server-side to all agent-driven downloads, blocking malicious packages, unapproved licenses, and policy-violating packages regardless of whether session steering or `jf setup` is active. • Covers indirect and transitive installs — postinstall scripts, `pip install -r requirements.txt`, `npm ci`, and equivalent bulk-pull commands — routing all through Artifactory from the first session. • Produces complete audit trails for every agent-driven download, block, and policy decision alongside existing human-driven build logs. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/jfrog https://jfrog.com/blog/platform-federation-syncs-multisite-development/ https://jfrog.com/blog/platform-federation-syncs-multisite-development/ Mon, 24 Aug 2026 15:30:33 GMT JFrog Platform Federation introduces a peer-to-peer mesh that syncs projects, access control, curation policies, and repositories across every site automatically. • Adds Projects Federation, which synchronizes multi-tenant workspaces, project- and global-level roles, SDLC release-stage gates, and identity access tokens across every peer site in the mesh in seconds. • Adds Curation Federation, which propagates open-source blocking rules, waivers, and approvals from any peer site to every regional pipeline worldwide within seconds of being set. • Adds Automated Repository Provisioning, which creates and configures repositories on the correct sites automatically based on a Stage-to-Site mapping, eliminating per-repo and per-site manual administration. • Introduces a peer-to-peer mesh architecture for distributed JFrog sites, replacing hub-and-spoke and script-driven sync with autonomous, continuous background synchronization of projects, access control, policies, artifacts, and repositories. • Supports both self-managed and JFrog Cloud SaaS sites — including airgapped and private-cloud deployments — in the same federated mesh, enabling phased SaaS migrations without breaking centralized security or identity control. • Provides a Federated Topology View showing the global peer mesh across regions with entities and policies synchronized in real time. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/jfrog https://jfrog.com/blog/amazon-bedrock-agentcore-gateway-jfrog-artifactory/ https://jfrog.com/blog/amazon-bedrock-agentcore-gateway-jfrog-artifactory/ Fri, 21 Aug 2026 16:20:31 GMT JFrog Artifactory gains On-Behalf-Of (OBO) OAuth 2.0 token exchange with Amazon Bedrock AgentCore Gateway for per-user AI agent identity propagation. • Supports On-Behalf-Of (OBO) token exchange (RFC 8693) between Amazon Bedrock AgentCore Gateway and JFrog Artifactory Access, so AI agent requests reach Artifactory authenticated as the originating user rather than a shared service account — preserving audit fidelity, least privilege, and containment. • Enables AgentCore Gateway to use JFrog Artifactory's MCP server as a registered target, with `onBehalfOfTokenExchangeConfig` configured on the gateway target so each inbound user JWT is exchanged for a short-lived, per-user JFrog token. • Supports `listingMode` field on AgentCore Gateway targets (requires a recent AWS CLI) to enable dynamic tool listing, allowing tool discovery to run under the user's token instead of a shared credential. • Integrates JFrog Access OIDC identity mappings with AgentCore Identity: JFrog trusts the external IdP issuer (`$IDP_ISSUER`) and maps verified access-token claims (e.g. `preferred_username`, `team`) to JFrog users, groups, or project roles via the JFrog OIDC integration API. • Requires the user's OIDC access token `aud` claim to equal the AgentCore Gateway URL (`$GATEWAY_URL`) and `iss` to match `$IDP_ISSUER`, with a mappable identity claim (e.g. `preferred_username`) present on the access token — not only the ID token. • Compares three identity-propagation options across the gateway-to-Artifactory hop: header whitelisting (forwards `X-Authenticated-User`-style headers), three-legged OAuth (authorization-code grant with per-user token vault in AgentCore Identity), and On-Behalf-Of exchange — with OBO being the only option that delivers verifiable per-user identity at Artifactory without a second login. • Requires JFrog EnterpriseX or higher for OIDC token exchange support on the Artifactory Access service. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/jfrog https://jfrog.com/blog/amazon-bedrock-agentcore-gateway-jfrog-artifactory/ https://jfrog.com/blog/amazon-bedrock-agentcore-gateway-jfrog-artifactory/ Fri, 21 Aug 2026 16:20:31 GMT JFrog Artifactory gains On-Behalf-Of (OBO) OAuth 2.0 token exchange with Amazon Bedrock AgentCore Gateway for per-user AI agent identity propagation. • Supports On-Behalf-Of (OBO) token exchange (RFC 8693) between Amazon Bedrock AgentCore Gateway and JFrog Artifactory Access, so AI agent requests reach Artifactory authenticated as the originating user rather than a shared service account — preserving audit fidelity, least privilege, and containment. • Enables AgentCore Gateway to use JFrog Artifactory's MCP server as a registered target, with `onBehalfOfTokenExchangeConfig` configured on the gateway target so each inbound user JWT is exchanged for a short-lived, per-user JFrog token. • Supports `listingMode` field on AgentCore Gateway targets (requires a recent AWS CLI) to enable dynamic tool listing, allowing tool discovery to run under the user's token instead of a shared credential. • Integrates JFrog Access OIDC identity mappings with AgentCore Identity: JFrog trusts the external IdP issuer (`$IDP_ISSUER`) and maps verified access-token claims (e.g. `preferred_username`, `team`) to JFrog users, groups, or project roles via the JFrog OIDC integration API. • Requires the user's OIDC access token `aud` claim to equal the AgentCore Gateway URL (`$GATEWAY_URL`) and `iss` to match `$IDP_ISSUER`, with a mappable identity claim (e.g. `preferred_username`) present on the access token — not only the ID token. • Compares three identity-propagation options across the gateway-to-Artifactory hop: header whitelisting (forwards `X-Authenticated-User`-style headers), three-legged OAuth (authorization-code grant with per-user token vault in AgentCore Identity), and On-Behalf-Of exchange — with OBO being the only option that delivers verifiable per-user identity at Artifactory without a second login. • Requires JFrog EnterpriseX or higher for OIDC token exchange support on the Artifactory Access service. Full history on Cyber Toolchain: https://cybertoolchain.io/tools/jfrog