task-to-pr

Pass

Audited by Gen Agent Trust Hub on Jul 2, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted external data from ticket IDs, URLs, descriptions, and comments. This creates a risk of indirect prompt injection where malicious instructions embedded in a ticket could influence the agent's subsequent actions, such as writing malicious code or performing unauthorized git operations.
  • Ingestion points: The agent reads ticket content, user comments, and linked context like design docs or specifications (SKILL.md).
  • Boundary markers: The instructions do not specify any delimiters or warnings to ignore embedded instructions within the ingested ticket data.
  • Capability inventory: The agent has the ability to implement code changes, branch, commit, push, and open pull requests (SKILL.md).
  • Sanitization: There is no mention of sanitizing or validating the ticket content before it is used to guide the implementation and review process.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 2, 2026, 11:05 PM
Security Audit — agent-trust-hub — task-to-pr