work-tracking

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill describes a workflow that ingests plan data from repository files and promotes it to external tracking systems, which constitutes a surface for indirect prompt injection from untrusted repository content. \n
  • Ingestion points: The agent is instructed to read multi-phase plans from locations like roadmap/, docs/plans/, and .agents/plans/ in the repository as specified in SKILL.md. \n
  • Boundary markers: Absent; the instructions do not include specific delimiters or 'ignore instructions' warnings for the ingested plan data. \n
  • Capability inventory: The skill directs the agent to perform file system writes and create issues in external platforms like GitHub and Linear. \n
  • Sanitization: No explicit content validation or escaping is described for processing plan data.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 09:54 AM
Security Audit — agent-trust-hub — work-tracking