Skip to content

Releases: S3DFX-CYBER/Claude-Skills-Security

Release list

v1.1.0 — 2026-04-20

v1.1.0 — 2026-04-20 Latest
Latest

Choose a tag to compare

View all tags
@coderabbitai coderabbitai released this 21 Apr 08:12
v1.1.0
b0b205b
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

What's Changed

Added

  • AGENTS.md — Agentic discovery metadata, invocation contracts, and GitHub MCP workflow for autonomous triage across full repositories
  • Multi-agent compatibility — Explicit installation and invocation guidance for OpenAI Codex, Cursor, Windsurf, Gemini CLI, and GitHub Copilot
  • COMPATIBILITY.md — Runtime-specific configuration details
  • **.claude/skills/ structure** — Native Claude Code agentic skill path
  • Agentic Mode section in skill file — GitHub MCP proactive fetch logic, SHA-pin auditing grep pattern, graceful degradation when MCP unavailable
  • CVSS agent guidance — Common agent CVSS errors documented (XSS S:C, CI/CD UI:R, transitive dep downgrade, AV:N context adjustment)

Changed

  • Track D (CI/CD) — Expanded pull_request_target detection heuristics with real-world PoC reference (CVE-2026-40316, GHSA-wxm3-64fx-cmx9)
  • Track C (LLM) — LLM01–LLM10 detection heuristics updated to OWASP LLM Top 10 2025
  • Confidence gate — False-positive checklist expanded with LLM model-level vs application-level distinction as mandatory pre-classification step
  • CVSS construction — Component-by-component reasoning now required in all reports; base score context adjustment added as mandatory field in AM I AFFECTED section

Fixed

  • LOW confidence issues were appearing in findings summaries in edge cases — gate logic now explicitly suppresses them to Notes section only

Full changelog: https://github.com/S3DFX-CYBER/Claude-Skills-Security/blob/main/CHANGELOG.md

Assets 2
Loading