woostack-fix
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon untrusted data from external platforms including Sentry, GitHub, Linear, and Plane.
- Ingestion points: The primary command
/woostack-fixaccepts<goal-or-untrusted-input>which can originate from external monitoring or issue tracking systems (SKILL.md). - Boundary markers: The instructions explicitly define external inputs as "untrusted evidence" and require the agent to prove a causal root cause before any repository mutation (SKILL.md, Phase 1).
- Capability inventory: The skill possesses the capability to perform repository mutations (via PRs) and interact with external project management APIs (Linear, Plane, GitHub) after proof is established.
- Sanitization: Instructions mandate independent verification of cited immutable source/diff identity and validation of writable owning checkouts (SKILL.md, Phase 2).
- [EXTERNAL_DOWNLOADS]: The skill utilizes official integrations and tools from well-known technology services to manage project state and repository access.
- Evidence: The instructions reference using the official GitHub CLI (
gh) and the Model Context Protocol (MCP) for interactions with Linear and Plane (SKILL.md, Project-backed Fix). - Note: These interactions target well-known developer platforms and follow established authentication and integration patterns for development workflows.
Audit Metadata