article-prompts-to-skills
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's core function is to ingest and process untrusted external data, such as articles and tutorials, to generate new agent instructions and capability packages. This workflow presents an indirect prompt injection surface where malicious content in the source article could influence the behavior of the generated skills.
- Ingestion points: Source articles, tutorials, and prompt packs are read in Step 1 of the extraction process (SKILL.md).
- Boundary markers: No specific boundary markers or 'ignore' instructions are mandated for the ingestion of the source text.
- Capability inventory: The skill utilizes file system writes, Git operations (status, diff, commit), and execution of a local validation script (
quick_validate.py). - Sanitization: The skill includes a specific validation step (Step 7.8) to scan for secrets, tokens, and private paths before any changes are committed to the repository.
- [COMMAND_EXECUTION]: The skill directs the agent to perform shell-based repository management and validation tasks.
- Evidence: Instructions include running
git status --short(SKILL.md, Step 1.2),quick_validate.py(SKILL.md, Step 7.1), andgit diff --cached(SKILL.md, Step 8). These commands are consistent with the skill's purpose as a developer tool.
Audit Metadata