The daily firehose — everything the toolchain shipped today, already filtered.
We read every release from the 1309 tools on our watchlist at the source — GitHub and GitLab release notes, vendor release pages and changelogs, project blogs and feeds, vendor press releases, and the source code behind the tag. Bug-fix-only releases and non-product newsroom noise are dropped; what's left is summarized down to the new capability, how to try it, and any screenshots or videos the release itself published. Every entry links to the sources it was built from.
READ AS
Focus:What features shipped and what problems they solve, plus launches, packaging and where the market is moving.
Leaves out:Command-line flags, API-only deltas, usage examples, and releases that are only refinements.
Focus:New security controls, new APIs, and product improvements you can put to work, plus anything an upgrade breaks.
Leaves out:Launch and marketing news, governance, pricing, and documentation changes with no operational effect.
Focus:Dashboards, reporting, and anything you can quantify: metrics, coverage, risk scores, compliance impact, vendor moves.
Leaves out:Settings, flags, API and command-line detail, library bumps, and minor releases with no reporting or posture angle.
Focus:API, SDK and MCP changes, new libraries, breaking changes in what you import, and cyber tooling.
Leaves out:Governance, SOC operations, and vendor launches with no developer surface.
AMOUNT
DETAIL
ORDER
This issue came out before editions were ranked, so it shows the Standard edition in issue order.
Codacy is an automated code quality and security analysis platform that reviews pull requests, enforces coding standards, and tracks technical debt across repositories.
Codacy now respects ShellCheck configuration files in your repository.
Portkey is an AI gateway and observability platform that helps developers manage, monitor, and optimize LLM API calls with routing, caching, and reliability features.
Portkey adds vault-backed secret references, weekly/endpoint-scoped rate limits, two new guardrail integrations, and broader provider coverage.
└──▷ USE IT
Apply Zscaler Detection Policies to all 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. traffic by wiring Zscaler AI Guard into your Portkey config as a guardrail hook.
Set a weekly spend cap scoped only to the embeddingsembeddingsLists of numbers a model produces to represent the meaning of text, images or code, so similar items sit close together. They power semantic search and RAG. endpoint so it doesn't share budget with chat completions.
📍In the Portkey console, go to Budget & Rate Limit Policies, create a new policy, set the window to rpw (weekly), set the endpoint scope to embeddings, and enter your desired token or cost limit.
Introduces weekly rate/budget windows (rpw) and endpoint-scoped rate limits, letting teams cap spend per API surface (chat completions, embeddings, admin routes, etc.) independently.
Veracode is a cloud-based application security platform that helps developers identify and remediate vulnerabilities in their code through static, dynamic, and software composition analysis.
Veracode integrations gain platform-driven GitHub install/uninstall, scheduled scans, new API pagination, and broader language autopackaging support.
└──▷ USE IT
Enable debug logging for a GitHub or Azure DevOps workflow scan to troubleshoot packaging or scan failures.
yaml
# In veracode.yml
debug: true
Install and schedule Veracode GitHub Integration scans from the platform without touching the GitHub Marketplace.
📍In the Veracode Platform, go to Integrations › GitHub › Install, select your organization, configure credentials, then go to Schedule Scans to set up Static Analysis, SCA, or IaC scan schedules.
Scope a Visual Studio scan to only the active solution for faster, more focused results in large repos.
📍In Visual Studio with the Veracode Scan extension, open Settings › Veracode Scan and set Scan Scope to Active Solution, then run a scan.
Adds a new uninstallation workflow on the Veracode Platform to remove the Veracode app from GitHub organizations.
Enables installing, configuring, and updating credentials for the Veracode GitHub Integration directly from the Veracode Platform, bypassing manual GitHub Marketplace steps.
Adds autopackaging support for Apex, VB6, and Perl in the GitHub Workflow Integration.
Enables scheduling Static Analysis, 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 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. scans from the Veracode Platform for GitHub organizations using the GitHub Workflow Integration.
Adds scheduled scan support for every organization where the Azure DevOps Workflow Integration is installed, driven by veracode.yml configuration.
+14 moreshow less
Adds a Scan Scope setting in Veracode Scan for Visual Studio to scan the full project directory or only the active solution.
Adds Visual Studio 2026 support in Veracode Scan for Visual Studio.
Adds page size support to the Platform Applications API (/appsec/v1/applications) in the Java API Wrapper for efficient large-list retrieval.
Introduces support for the Veracode Platform paginated Sandbox 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. in the Java API Wrapper, enabling pagination-based sandbox list retrieval.
Adds support for the Veracode Platform Findings REST API in the Java API Wrapper to access open and mitigated application findings.
Adds a configurable retry mechanism for the Veracode Platform Applications API in the Java API Wrapper.
Adds a debug configuration option in veracode.yml for the Azure DevOps Workflow Integration, GitHub Workflow Integration, and GitLab Workflow Integration for debugging and troubleshooting.
Adds support for Windows-based runners in the GitHub Workflow Integration.
Enforces custom platform policies with Build Action: error for SCA agent-based scans in the Azure DevOps Workflow Integration, failing the scan when vulnerabilities are found.
Imports static scan results from the Veracode Platform directly into the Visual Studio IDE.
Proposes mitigations for static scan flaws directly within the Visual Studio IDE.
Ensures findings marked as mitigated on the platform are excluded from Azure DevOps pipeline scan results.
Adds usage analytics in Veracode Scan for JetBrains, VS Code, and Visual Studio for insights into plugin adoption, feature usage, and scan activity.
Scheduled scans in the Azure DevOps Workflow Integration now run against the analysis_branch specified in veracode.yml instead of the default branch.
19 details in Everything
└──▷ BREAKING ON UPGRADE
!The Veracode Flaw Importer task (with spaces in the task name) in the Azure DevOps extension is deprecated and replaced by VeracodeFlawImporter; the old task is scheduled for removal in a future release.
2 more releases in this issue
· 2026-07-29 → 2026-07-30
Veracode adds Custom PDF 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., OAuth for XML APIsXML APIsInterfaces for reading, writing, or querying XML documents, provided by libraries such as libxml2 or built into language runtimes; cyber tools use them to parse structured data like configs, reports, or protocol messages reliably., six new built-in policies, and expanded Reporting API filters in its 2026 platform updates.
New Custom PDF REST API enables programmatic generation of custom PDF reports.
XML APIs now support OAuth Client Credentials authentication, in addition to REST APIsREST APIsA convention for building HTTP-based interfaces where resources are addressed by URL and manipulated with standard verbs like GET, POST, and DELETE, giving cyber tools a predictable way to integrate with external services..
Auto-Update 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. and CWECWECommon Weakness Enumeration, a catalog of recurring software and hardware weakness types. It describes classes of mistakes, such as injection, rather than individual vulnerability instances. Top 25 policy rules now enforce OWASP Top 10:2025OWASP Top 10:2025A ranked list of the ten most critical web application risk categories, maintained by the Open Worldwide Application Security Project, used by cyber tools to prioritize detection, testing, and remediation coverage. and MITREMITREA US federally funded research and development center that maintains public frameworks—ATT&CK, CVE, D3FEND—that cyber tools use to classify threats, vulnerabilities, and defensive techniques in a common language. CWE Top 25 for 2025 standards.
Six new built-in security policies added to the Veracode Platform.
Veracode Analytics upgraded to Looker 26.6 with finer widget sizing, persistent filtering, and tabbed dashboards.
+18 moreshow less
Finding Reporting API responses now include a collections_list field returning collection names associated with an application.
Veracode Analytics Applications explore now includes Collections fields, and the Policy Compliance Overview dashboard adds Collections visualizations.
Reporting API Scans report supports new most_recent_scan filter to return only the most recent scan per asset.
OAuth 2.0OAuth 2.0An open authorization framework (IETF RFC 6749) that lets a service grant limited API access to a third party without sharing credentials, giving cyber tools a standard way to authenticate against external platforms. Client Credentials can now be created and managed directly on the API Credentials page in the Veracode Platform.
New Firewall Usage Dashboard and Explore in Veracode Analytics surface metrics for Veracode Firewall.
Policy Settings page now supports light and dark mode.
New library_instance_ref field added to Reporting API findings responses to uniquely identify specific library instances.
Reporting API Findings report supports new custom_severity filter.
New data_refresh_date field added to Reporting API metadata indicating the most recent Analytics data refresh time.
Manual Penetration Testing (MPT) Flaw Matching correlates findings across tests for consistent flaw IDs, status values, and grace period inclusion.
Administrators can now disable Veracode-defined roles for users who use custom roles, via the Platform UI and updated Identity API.
Container and 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. scan results from the CLI are now viewable on the Veracode Platform, including severity counts, filtering, scan history, and mitigation guidance.
New action_type filter available in the Reporting API Audit Log report for narrowing records by specific user actions.
Reporting REST API now supports generating an Audit Log report for querying detailed user activity.
Multiple domains can now be mapped to a single identity provideridentity providerA service that authenticates users and supplies identity information to other applications. It is the sign-in authority behind many single sign-on systems. for SP-initiatedSP-initiatedA SAML or OAuth/OIDC login flow where the Service Provider redirects the user to an Identity Provider to authenticate, rather than the IdP pushing an assertion unprompted; tools use it to trigger SSO from their own login page.SAML SSOSAML SSOAn authentication protocol that lets a user log in once with an identity provider and access multiple services without re-entering credentials, by passing signed XML assertions between parties..
Customizable PDF reports now display 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. Component License Risk policy compliance in Executive Summary and Detailed SCA Findings sections.
Administrators can now customize columns in the All Users table on the Administration page.
Customer Managed Encryption Keys (CMEK) allows encryption of Veracode Platform data using an organization's own root encryption key, with KMSKMSA managed service that creates, stores, rotates and controls access to encryption keys, usually backed by HSMs. Applications ask it to encrypt or decrypt rather than holding keys themselves. management and key rotation support.
23 details in Everything
└──▷ BREAKING ON UPGRADE
!The Veracode Healthcheck endpoint /healthcheck/status has been removed and now returns a 404 error; use the v2/principal endpoint in the Identity REST API instead.
Veracode SASTSASTStatic application security testing, analyzing source code or compiled code for vulnerabilities without running the application. It helps developers find problems before deployment. adds Pipeline Scan support for >200 MB artifacts, exploit-level scoring in results.json, and an enhanced Python scanner with PySpark/Flask/Django/FastAPI support.
Pipeline Scan now supports packaged artifacts larger than 200 MB with no integration changes required.
Pipeline Scan results.json now includes a numeric exploit level (-2 to 2) for each finding, surfacing exploitability likelihood alongside vulnerability data.
Static Analysis Engine v2026.07 adds support for Android 17 and enhances .NET 10 support.
Static Analysis Engine v2025.09 adds support for .NET Semantic Kernel InvocationFilter and Filter.
Static Analysis Engine v2025.09 adds support for Dart 3.9, Flutter 3.35, and JDK 25 (LTS).
Static Analysis Engine v2025.09 updates Gen IR: the --capture flag now includes diagnostic data in .xcarchive files.
Adds initial support for iOS 26 in static analysis.
Static Analysis Engine v2025.08 adds .NET Semantic Kernel support for detecting CWECWECommon Weakness Enumeration, a catalog of recurring software and hardware weakness types. It describes classes of mistakes, such as injection, rather than individual vulnerability instances.-798 and CWE-15.
Static Analysis Engine v2025.08 adds support for Azure Functions v4 on .NET 9 in isolated process mode.
Static Analysis Engine v2025.08 adds support for Go 1.25, Node.js 22.x, Scala 3.7, Dart 3.8, and Flutter 3.32.
Static Analysis Engine v2025.08 adds support for Micronaut 4.x and improves Quarkus support for Java.
Introduces initial support for Android 16.
Adds support for Apex 63 and improves detection of SOSL FIND queries.
Infisical is the open-source platform for secrets, certificates, and privileged access management.
Infisical v0.162.15 adds SCEPSCEPSimple Certificate Enrollment Protocol, an IETF-standardized protocol for automating X.509 certificate issuance and renewal over HTTP, letting cyber tools request PKI certificates from a CA without manual intervention. support for Microsoft Intune, NKP Helm chart, PAMPAMBrokering administrator sessions through a gateway instead of handing out standing credentials. Access is granted for one session, to one target, and recorded. connection testing, and improved machine identity template visibility.
└──▷ GET THIS VERSION
$ git clone --branch v0.162.15 https://github.com/Infisical/infisical.git
# already have the repo? check out this version:$ git checkout v0.162.15
Adds SCEP (Simple Certificate Enrollment Protocol) support for Microsoft Intune, enabling certificate-based device enrollment flows.
New Helm chart (infisical-nkp) for deploying Infisical in Nutanix Kubernetes Platform (NKP) partner catalog environments.
Adds connection test capability for PAM (Privileged Access ManagementPrivileged Access ManagementBrokering administrator sessions through a gateway instead of handing out standing credentials. Access is granted for one session, to one target, and recorded.) accounts, allowing validation of account connectivity before use.
Improves machine identity template visibility in the UI, making templates easier to discover and manage.
GitGuardian Detection Engine scans code repositories and developer environments to automatically detect secrets, credentials, and sensitive data exposures in real time.
GitGuardian Detection Engine 2.168 adds 4 new detectors, 2 new validity checkers, and extends HTML/Markdown scanning to 5 token types.
New detector for Azure Immersive Reader Key credentials.
New detector for Alibaba Cloud Model Studio (Qwen / DashScope) Coding Plan API keys.
New detector for Alibaba Cloud Model Studio (Qwen / DashScope) Token Plan API keys.
New detector for Prefect Cloud API keys.
New validity checker for Alibaba Cloud IAMIAMIdentity and Access Management — a framework of policies and services for controlling who can authenticate to systems and what resources they are authorized to use, letting cyber tools enforce least-privilege and audit access. keys.
+4 moreshow less
New validity checker for GitLab Deploy Tokens when username and host are present.
Generic High Entropy Secret detector now covers additional credential key-name prefixes: access, app, vault, sign, dev, prod, and staging.
Okta Token with Host detector updated to match regional and preview host domains, raising precision from ~0% to ~100% on reviewed samples.
HTML and Markdown file scanning extended to npm, GitHub Personal Access, GitHub OAuth, GitHub App, and Slack Application tokens.
Security Onion is a free and open platform for threat hunting, enterprise security monitoring, and log management. It includes our own interfaces for alerting, dashboards, hunting, PCAP, detections, and case management.
Security Onion 3.2.0 adds agentic AI (Gemma-backed Investigator/Engineer), Sigma playbooks, ES|QLES|QLElasticsearch Query Language, maintained by Elastic, a pipe-based query language for filtering, transforming, and aggregating data stored in Elasticsearch, giving cyber tools a concise syntax for log and event analysis. defaults, data stream lifecycle management, and UEK8 kernel support.
└──▷ GET THIS VERSION
$ git clone --branch 3.2.0-20260729 https://github.com/Security-Onion-Solutions/securityonion.git
# already have the repo? check out this version:$ git checkout 3.2.0-20260729
Adds agentic AI capability with Gemma-backed Investigator and Detection Engineer agents for automated analysis workflows.
Supports Sigma playbooks (including airgap environments), enabling SigmaHQSigmaHQAn open community repository on GitHub that maintains the Sigma generic signature format and its growing library of detection rules, giving cyber tools a vendor-neutral rule set they can convert to any SIEM query language. rules against Elastic Defend logs.
Adds data stream lifecycle management support for Elasticsearch index management.
Upgrades to Elasticsearch 9.3.7.
Adds UEK8 kernel support across install and grid, with UEK8 set as the boot default.
+4 moreshow less
Adds cluster health monitoring option to telegraf.conf for Elasticsearch observability.
Makes failed soup upgrades and hotfixes resumable instead of requiring a full restart.
Adds toggleable PostgreSQL metrics collection.
Supports multiple capinfos versions for PCAPPCAPA packet capture, recorded network packets that can be inspected later. Analysts use it to reconstruct communication and investigate protocol behavior or suspicious traffic. analysis compatibility.
9 details in Everything
└──▷ BREAKING ON UPGRADE
!The icsnpp-modbus parser is removed from defaults.yaml; grids relying on the default Modbus protocol parser will no longer have it enabled after upgrade.
!The Detection Engineer agent configuration key is renamed to DetectionEngineer; existing configurations using the old key will need to be updated.
Fastly is a cloud edge platform that accelerates, secures, and delivers websites, apps, and APIs by caching content closer to users globally.
Next-Gen WAFWAFA reverse proxy that inspects HTTP/HTTPS traffic and blocks requests matching known attack patterns such as SQL injection or XSS, sitting between clients and a web application to filter malicious input before it reaches the server. Apache module now logs failed-open requests for better visibility into bypass events.
Logs failed-open requests, giving practitioners visibility into traffic that bypassed WAF inspection.
Sysdig Secure is a cloud-native security platform that detects threats, manages vulnerabilities, ensures compliance, and responds to incidents across containers, Kubernetes, and cloud environments.
Sysdig CLI Scanner v1.28.0 adds NixOS and Arch Linux package detection plus dnf/microdnf support.
Adds detection of NixOS (nix) packages in the CLI Scanner.
Adds detection of Arch Linux packages in the CLI Scanner.
Adds dnf and microdnf to the package manager detection policy rule.
Trend Vision One is a unified cybersecurity platform that integrates extended detection and response (XDR) across endpoints, networks, email, and cloud environments.
Container Security inventory gains resource-level drill-down, 24-hour trend indicators, and direct log links.
└──▷ HOW TO FIND IT
Investigate whether a specific ECSECSElastic Common Schema, an open specification maintained by Elastic that defines a standard field naming convention for log and event data, letting cyber tools share and correlate records without per-source field mapping. task is generating more findings than yesterday without leaving the inventory view.
📍In the console, go to Cloud Security → Container Security → Inventory, select an ECS service or task, then review the 24-hour trend indicator next to each finding type.
Jump directly from a finding count to the underlying log events for a specific Kubernetes pod.
📍In the console, go to Cloud Security → Container Security → Inventory, select a pod, then select a finding count to open the corresponding log page.
Drill down to individual Kubernetes nodes/pods or Amazon ECS services/tasks in the Container Security inventory overview, scoping all security findings to that single resource.
Each security finding type now displays a 24-hour trend indicator comparing the last 24 hours against the previous 24 hours.
Selecting a finding count in the inventory overview opens the corresponding log page directly for event-level review.
DefectDojo 3.1.302 adds login rate limiting to API token auth and expands scanner protocol support.
└──▷ GET THIS VERSION
$ git clone --branch 3.1.302 https://github.com/DefectDojo/django-DefectDojo.git
# already have the repo? check out this version:$ git checkout 3.1.302
Applies login rate limiter to the API token authentication endpoint, reducing brute-force exposure on token-based logins.
Adds additional scanner protocols to the whitelist, broadening supported import sources.
Updates Fortify parsers to extract and populate CWECWECommon Weakness Enumeration, a catalog of recurring software and hardware weakness types. It describes classes of mistakes, such as injection, rather than individual vulnerability instances. data from scan results.
API now returns HTTP 409 instead of 500 on unique-constraint violations, enabling cleaner client-side error handling.
runZero is a continuous asset intelligence platform that discovers, inventories, and monitors all connected devices and software across networks in real-time.
Immuta is a data security platform that automates access control, privacy policy enforcement, and compliance monitoring across cloud data environments.
Expands automated sensitive data discovery (regex, dictionary, and column-name regex identifiers) to Immuta's Amazon Redshift viewless integration.
Snowflake integration error messages are moved out of error banners and into the connections experience.
└──▷ BREAKING ON UPGRADE
!Immuta policy role names used to grant users access to Snowflake data sources will be renamed as part of the Snowflake integration architecture update rolling out from July 23, 2026.
DigiCert CertCentral is a centralized platform for managing SSL/TLS certificates, automating issuance, renewal, and deployment across organizational infrastructure.
CertCentral automated delegation approval expands to EV TLS, EV Code Signing, Verified Mark, and PKIo products globally and in Europe.
└──▷ HOW TO FIND IT
Delegate approval authority to a CertCentral user so future EV TLS or EV Code Signing orders are issued automatically without manual intervention.
📍In CertCentral, go to Account › Users. Select the target user. On the user details page, under Order approval delegation, select Request delegation. In the Submit a delegation request window, select the authorized representative or verified contact, then select Submit Request.
Extends automated delegation approval to EV TLS, EV Code Signing, Code Signing, Verified Mark, and Common Mark certificates in CertCentral global and Europe.
Adds automated delegation approval support for a broad set of CertCentral Europe products including EU Qualified eSeal (PSD2), EU Qualified Website Authentication Certificate (PSD2), PKIo Personal Organisation, PKIo Qualified eSeal, PKIo Private Person, and multiple PKIo Authentication/Encryption and Private Services Server variants.
Simplifies the delegation request workflow: a single request now covers all organizations the selected approver represents and all certificate products they are authorized to approve, removing the previous per-organization selection step.
Enables fully automated certificate issuance via CertCentral UI and Services API once a delegated approver is assigned, eliminating repeated manual approvals per order.
Nudge Security discovers and secures all SaaS apps used across an organization, helping IT teams manage shadow IT and enforce security policies through automated nudges.
Nudge Security AI conversation monitoring now detects sensitive data entered into Amazon Quick Suite chats.
└──▷ HOW TO FIND IT
Enable detection of sensitive data shared in Amazon Quick Suite for your organization.
📍In the Nudge Security console, go to Browser Extension Settings and configure AI conversation monitoring to include Amazon Quick Suite.
Detects secrets, 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., financial data, and health data shared in Amazon Quick Suite (Amazon's agentic AI platform) via AI conversation monitoring.
Nudge Security now discovers AI agents built on Amazon Quick, surfacing creator, components, and risk insights in your AI agents inventory.
└──▷ HOW TO FIND IT
Enable browser-based agent discovery to start automatically surfacing Amazon Quick AI agents in your inventory.
📍In the browser extension settings, turn on browser-based agent discovery. Then navigate to Identities › AI Agents tab in Nudge Security to view discovered Amazon Quick agents with their creator, components, and risk insights.
Mondoo is a security platform that continuously scans and assesses infrastructure, cloud environments, and software supply chains for vulnerabilities and misconfigurations.
Mondoo launches a hosted 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, enabling Claude Desktop, Claude Code, and VS Code to query spaces, findings, and policies in plain language.
New hosted MCP server lets AI tools (Claude Desktop, Claude Code, VS Code) connect to Mondoo using a read-only API token.
Enables natural-language exploration of Mondoo spaces, findings, and policies directly from AI-powered clients.
Windsurf is an AI-powered code editor that combines an intelligent coding assistant with agentic capabilities to help developers write, edit, and debug code faster.
Windsurf v3.6.21 adds browser preview for remote ACP agents, Cascade Migration Wizard, Plan mode for Devin Local, and editable command approvals.
└──▷ HOW TO FIND IT
Migrate all Windsurf hooks, skills, and memories to Devin in one guided flow with dry-run previews before committing any changes.
📍Open the Command Palette (⌘P / Ctrl+P) → run Devin: Open Cascade Migration Wizard → review each hooks/skills/memories item with per-item opt-in and dry-run preview → confirm to apply.
Suppress Generated by Devin attribution lines from commit messages produced by Devin Local in your repository.
yaml
# In your Devin Local config file:
attribution: false
Adds a unified notifications setting that posts native OS notifications when any agent session finishes or needs input.
Codemaps now open in their own editor tabs in agent window mode, and codemap @-mentions now work in Devin Local.
New Add to chat button (⌘L / Ctrl+L) lets users select text in a message transcript and append it to chat input.
Adds keyboard shortcuts for permission request actions (always-allow and reject) with shortcut hints shown on buttons.
Pull request editor tabs now show state-specific icons (open, draft, merged, or closed); session-create cards show a Devin mode badge (Fast, Ultra, Fusion).
+21 moreshow less
Remote ACP sessions can now open a browser preview, proxying the local dev server with captured elements and console output sent as pending context.
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 reporting Needs auth now show an Authenticate button that clears stored OAuth credentials and reruns the browser authorization flow.
New Devin: Open Cascade Migration Wizard command chains hooks → skills → memories migration in one guided flow with per-item opt-in and dry-run previews.
Plan mode for Devin Local: agent researches with read-only commands, persists a Markdown plan file at ~/.devin/plans/plan-<session>.md, and asks for approval before implementing.
Worktree sessions now show a Merge button to bring changes back into the workspace, with an existing-worktree picker in the agent location selector.
Editable command approvals: click a command in a permission card to edit it before approving, or use the wand action to rewrite it via a fast model.
Queued messages can now be edited before they send; editing or reverting a prompt preserves file, directory, skill, and terminal @-mentions.
Adds Rules to the @-mention menu so manual-trigger rules can be pulled into a conversation.
Adds a Duplicate session action to the response footer to branch off a conversation.
Adds a Subagents (Preview) toggle in Devin settings; custom subagents can be defined as flat agents/<name>.md files.
Devin Cloud sessions now show the server-side send queue in sync with the web app; Cmd/Ctrl+Enter queues the next message on the server.
Multi-user (Slack) sessions now show each participant's name above their messages.
Adds a Copy link action to cloud session menus.
Adds a Remote ACP is disconnected banner with a Reconnect action when the Cloud connection drops.
Session network policy can now be configured and network access requests granted or denied inline from the chat without switching to the web app (v3.5.17).
Hooks tab in Devin Local Customizations now lists configured hooks with their source and trigger events (v3.5.17).
Adds a command to migrate Windsurf hooks into Devin hooks across every workspace folder (v3.5.17).
Adds Fast Context support in Devin Local sessions (v3.5.17).
Adds sandbox.excluded allow/ask/deny config (user and team settings) to run specific commands outside the sandbox (v3.4.22).
Subagents can now be configured with a default model (v3.3.18).
Adds an attribution option to the Devin Local config file to suppress Devin mentions in commit messages (v3.3.18).
26 details in Everything
└──▷ BREAKING ON UPGRADE
!The status bar is now hidden by default in Agent Command Center mode; set "workbench.statusBar.visible": true in user settings to restore it.
!Cascade, Devin Local, and every other ACP agent are unavailable while a workspace is open in Restricted Mode, and hooks no longer load or run there.
!Models marked Devin Local only (including every GPT-5.6 variant) now appear disabled in Cascade's model picker rather than running degraded.
Cursor is an AI-powered code editor that helps developers write, edit, and debug code faster using integrated large language model assistance.
Cursor launches native iPad app with split-screen reviews, Apple Pencil markup, Inbox, and Bitbucket/Azure DevOps support.
└──▷ HOW TO FIND IT
Open a PR review alongside an active agent chat on iPad to read diffs and prompt the agent to resolve comments without switching screens.
📍In the iPad app, open a PR from the Inbox, then use Split Screen to place the review alongside your active chat. Tap a comment and prompt the agent to resolve it directly from the review panel.
Annotate a screenshot with Apple Pencil to give the agent precise visual feedback on a UI issue.
📍In a chat, attach a screenshot, then tap a specific point on the image to drop a comment, or draw on it with Apple Pencil before sending to the agent.
Adds native iPad app (available on all paid plans) with a rebuilt layout featuring pinned sidebar chats for monitoring multiple agents simultaneously.
Supports split-screen mode on iPad, keeping a PR review open alongside an active chat with full file diff rendering.
Enables Apple Pencil markup on screenshots: tap to drop a comment at a specific point or draw directly on an image.
Adds a full PR review surface on iPhone and iPad covering comments, checks, approvals, reviewer management, and agent-driven resolution.
Introduces an Inbox to track in-progress work, items needing attention, and PRs currently in review.
+3 moreshow less
Adds Bitbucket and Azure DevOps 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. support on mobile.
Supports multi-PR sessions: when one chat creates multiple PRs, all of them can now be opened, not just the last.
Enables switching between teams directly within the mobile app.
Lightweight coding agent that runs in your terminal
Codex CLI adds 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. 2026 protocol support, cloud-managed sandbox profiles, persisted thread sections, and model-owned token budget defaults.
└──▷ GET THIS VERSION
$ git clone --branch rust-v0.147.0-alpha.2 https://github.com/openai/codex.git
# already have the repo? check out this version:$ git checkout rust-v0.147.0-alpha.2
Adds MCP 2026-07-28 discovery support and completes MCP 2026 client support, enabling use of the latest Model Context ProtocolModel Context ProtocolModel 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. spec.
Exposes cached MCP tools to inference before the MCP server finishes initializing, reducing startup latency for tool-heavy workflows.
Routes MCP OAuthMCP OAuthAn authentication layer for the Model Context Protocol that lets MCP clients obtain delegated access tokens so cyber tools can call MCP servers on a user's behalf without handling raw credentials. requests through configured HTTP clients, enabling proxy-aware MCP authentication.
Adds persisted sections for organizing threads in session history.
+11 moreshow less
Supports model-owned token budget defaults, letting model catalog entries supply default token-budget settings when none are explicitly configured.
Adds configurable developer instructions for v2 subagents.
Enables network policy callbacks for remote exec environments.
Routes WebRTCWebRTCA browser-native, open standard for real-time peer-to-peer audio, video, and data exchange without plugins. Cyber tools use it to tunnel traffic or exfiltrate data through connections browsers treat as legitimate. sideband joins to the Realtime API.
Supports self-serve Business ProLite accounts.
Supports streaming bodies in route-aware HTTP requests.
Supports plaintext collaboration tool messages.
Loads thread titles concurrently during session startup, improving startup responsiveness.
Tags turn reports with the selected turn's model and effort level.
Terminates Windows non-TTY processes on interrupt for cleaner process lifecycle management.
OpenHands v1.7.0 adds persistent agent memory, always-visible 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. selector, and secret overwriting from the edit form.
└──▷ GET THIS VERSION
$ git clone --branch v1.7.0 https://github.com/OpenHands/OpenHands.git
# already have the repo? check out this version:$ git checkout v1.7.0
Adds a persistent agent memory toggle so agents can retain context across sessions.
LLM selector is now always visible in the UI; agent-profile switching moves to the tools menu.
Enables overwriting an existing secret's value directly from the secret edit form.
Pi v0.83.0 adds credential export commands, headless OpenRouterOpenRouterA unified API gateway that routes requests to dozens of LLM providers under one endpoint, letting cyber tools swap or compare models without rewriting provider-specific integration code. login, and Claude Opus 5 via GitHub Copilot.
└──▷ GET THIS VERSION
$ git clone --branch v0.83.0 https://github.com/earendil-works/pi.git
# already have the repo? check out this version:$ git checkout v0.83.0
└──▷ TRY IT
Pipe a bearer tokenbearer tokenA credential that grants access to whoever presents it, without a separate proof of possession. Protecting the token matters because a stolen copy can be used by someone else. into another tool (e.g., a curl call to an API) without manually copying credentials.
Export an API key into an environment variable for a script that calls the provider directly.
$ export API_KEY=$(pi auth print-api-key)
Adds pi auth print-api-key and pi auth print-bearer-token commands to export configured credentials (with automatic OAuth refresh and minimum-validity enforcement) for use by external clients.
Enables headless OpenRouter login by accepting a manually pasted redirect URL or authorization code when the loopback callback is unavailable (e.g., over SSH).
Adds Claude Opus 5 support through the GitHub Copilot provider, with adaptive thinking and a 1 million-token context windowcontext windowThe maximum amount of text, counted in tokens, a model can take in and produce in one request. Everything the model knows about the task has to fit inside it..
Adds per-request fetch injection for supported text and image provider transports.
Adds the "pending" stop reason for partial streaming messages in the custom provider stream pattern.
5 details in Everything
└──▷ BREAKING ON UPGRADE
!TypeBox aliases upgraded to 1.3.7, removing deprecated APIs Type.Base, Type.Awaited, Type.Promise, Type.AsyncIterator, Type.Iterator, Type.Options, and Value.Mutate — extensions using any of these APIs must migrate to supported TypeBox APIs.
Autonomous coding agent as an SDK, IDE extension, or CLI assistant.
Cline CLI v3.0.47 adds free-model support, persistent settings, and agentic compaction as the new default.
└──▷ GET THIS VERSION
$ git clone --branch cli-v3.0.47 https://github.com/cline/cline.git
# already have the repo? check out this version:$ git checkout cli-v3.0.47
Adds support for free Cline models end-to-end, displayed with a (free) label and a dedicated card showing the reset time when the free usage limit is reached.
Persists /settings general toggles — plan/act mode, tool auto-approve, and compaction mode — across CLI restarts.
Expands built-in provider coverage, now generated from models.dev, with an updated bundled model catalog.
Framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks.
CrewAI 1.15.9 surfaces tool failures explicitly, emits FlowFailedEvent on flow errors, and adds progressive disclosure for skills.
└──▷ GET THIS VERSION
$ git clone --branch 1.15.9 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:$ git checkout 1.15.9
Surfaces tool failures as actual failures instead of silently reporting them as success, enabling agents to react correctly to tool errors.
Emits a new FlowFailedEvent when a flow execution fails, allowing event-driven error handling and observability hooks in Flows.
Adds progressive disclosure for skills, surfacing contextually relevant skill guidance incrementally rather than all at once.
A full-stack AI Red Teaming platform securing AI ecosystems via OpenClaw Security Scan, Agent Scan, Skills Scan, MCP scan, AI Infra scan and LLM jailbreak evaluation.
AI-Infra-Guard v4.5.1 adds multi-turn jailbreak attacks, 6 new 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. agent-scan skills, and 4 new 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. security detection rules.
└──▷ GET THIS VERSION
$ git clone --branch v4.5.1 https://github.com/Tencent/AI-Infra-Guard.git
# already have the repo? check out this version:$ git checkout v4.5.1
Adds Many-Shot, PAIR, GOAT, and ActorAttack multi-turn jailbreak attack methods to PromptSecurity.
Adds 5 new OWASP-aligned Agent-Scan detection skills: agentic-supply-chain, cascading-failure, human-agent-trust, inter-agent-comm, and unexpected-code-execution.
Adds web-exfiltration-detection skill to Agent-Scan.
Adds 4 new MCP-Scan security detection rules covering hardcoded secrets and insecure deserializationdeserializationTurning stored or transmitted data back into objects a program can use. Unsafe handling of untrusted data can allow unexpected behavior or code execution..
AgentMesh shifts to freemium with tiered API access, including a free signed-in API trial and paid tiers bundling Kirin endpoint licenses.
Introduces anonymous (no-login) free tier with access to 100 items per category via the Web UI.
Adds free signed-in tier unlocking 300 items per category plus trial API access (capped at 20 calls/month, 1/day) for engineers and security researchers.
Launches Premium tier ($15/user/month) with full catalog access, 300 API calls/month (20/day cap), and a bundled Kirin endpoint agent license.
Introduces Pro tier ($850/user/month) with unlimited AgentMesh API access and a Kirin IDE license for production pipelines and commercial integrations.
Adds Enterprise tier with custom pricing, dedicated support, and deployment for 10+ users.