fold-findings

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is potentially vulnerable to indirect prompt injection because it reads and processes data from external project files (e.g., review-findings.md). While it enforces a strict markdown table schema, it does not utilize explicit boundary markers or content sanitization to isolate this untrusted data from the agent's instructional context.\n
  • Ingestion points: The review-findings.md file located within project documentation paths.\n
  • Boundary markers: None specified for the findings ledger content.\n
  • Capability inventory: The skill can execute Git operations (commit, push) and project-specific verification tools (type-check, tests, build).\n
  • Sanitization: The instructions do not define methods for escaping or filtering content retrieved from the findings rows.\n- [COMMAND_EXECUTION]: The skill requires the agent to execute shell commands such as git commit and git push, and to run project-specific verification gates. These operations are essential to the skill's function but involve interacting with the system's command line and executing logic defined within the user's repository.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 11:08 PM
Security Audit — agent-trust-hub — fold-findings