inject-docs

Pass

Audited by Gen Agent Trust Hub on Oct 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches and executes the official Next.js codemod from Vercel's package registry using npx @next/codemod@canary to automate documentation injection.
  • [COMMAND_EXECUTION]: Executes shell commands via npx for Next.js and uv run for a local FastAPI injection script (inject_fastapi_docs.py).
  • [INDIRECT_PROMPT_INJECTION]: The skill modifies sensitive project files (CLAUDE.md or AGENTS.md) which define behavior for AI agents, creating a potential surface for indirect prompt injection.
  • Ingestion points: The skill reads package.json, pyproject.toml, and requirements.txt to detect frameworks and versions.
  • Boundary markers: The FastAPI script and Next.js codemod use specific headers to delimit the documentation sections being updated.
  • Capability inventory: The skill has access to Bash (npx, uv run, cat), Read, Grep, Glob, and AskUserQuestion tools, and possesses write capabilities for project documentation.
  • Sanitization: The skill does not explicitly sanitize framework version strings or other project metadata before injecting them into the project's agent instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 12, 2026, 05:09 PM