create-readme
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the repository's codebase and git history to generate documentation, which presents a surface for indirect prompt injection attacks.
- Ingestion points: The skill reads repository structure, configuration files (e.g.,
package.json,pom.xml,pyproject.toml), and git history (git log --oneline -n 50) in Phase 1. - Boundary markers: There are no explicit instructions or delimiters defined to separate untrusted repository content from the agent's internal instructions during the processing phase.
- Capability inventory: The agent has the ability to read arbitrary files within the repository, create new git branches (
feature/{YYYYMMDD}-readme-changelog), and commit changes to the repository. - Sanitization: The skill includes a manual verification step ('No secrets or credentials were accidentally included') but does not specify automated sanitization or filtering of the ingested repository content.
Audit Metadata