doc-engine
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill uses the
read_filetool to ingest repository content for documentation purposes, which presents a surface for indirect prompt injection from untrusted data within the codebase. - Ingestion points: The skill reads repository files, such as source code and existing documentation, using the
read_filetool. - Boundary markers: The agent is instructed to follow a strict output format and is gated by human approval requirements before overwriting critical files like
README.mdorAGENTS.md. - Capability inventory: The skill's scope is restricted to reading files and suggesting documentation edits. It does not request network access or unrestricted command execution.
- Sanitization: A validation script (
scripts/validate-output.sh) is provided to verify output structure and specifically prevent the inclusion of external remote resources (scripts/stylesheets) in generated HTML reports.
Audit Metadata