nightly-docs-sweep

Pass

Audited by Gen Agent Trust Hub on Aug 31, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes codebase content and documentation, which represents an attack surface for instructions embedded in data.
  • Ingestion points: Workflow steps 1 and 2 involve reading repository code, configuration, and documentation (SKILL.md).
  • Boundary markers: There are no explicit delimiters defined to separate untrusted codebase content from the agent's instructions.
  • Capability inventory: The skill is capable of writing to files, running project verification checks, and opening pull requests (Workflow steps 3, 5, and 7).
  • Sanitization: No specific sanitization or filtering of codebase content is mentioned before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 31, 2026, 12:39 AM
Security Audit — agent-trust-hub — nightly-docs-sweep