documentation-writer
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted repository content (source code, schemas, tests) to generate documentation, which creates a surface for indirect prompt injection.\n
- Ingestion points: Accesses and interprets source code, serializers, schemas, and tests as described in SKILL.md.\n
- Boundary markers: The skill lacks explicit instructions or delimiters to isolate repository content from its core instructions.\n
- Capability inventory: Executes shell commands via
uv run zensical serveanduv run zensical build(SKILL.md).\n - Sanitization: Relies on manual verification but does not define automated sanitization for content ingested from the codebase.\n- [COMMAND_EXECUTION]: The skill instructs the agent to run documentation management commands using the
uvpackage manager.\n - Evidence: Workflow steps involve running
uv run zensical serveanduv run zensical buildto process documentation.
Audit Metadata