skills/ferueda/harness/explain-change/Gen Agent Trust Hub

explain-change

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted external data such as code diffs and PR descriptions, creating an attack surface for indirect prompt injection.
  • Ingestion points: The agent resolves and reads code from branches, commits, and pull requests (SKILL.md sections 1 and 2).
  • Boundary markers: The instructions do not specify the use of delimiters or warnings to ignore instructions that might be embedded in the code or diffs being analyzed.
  • Capability inventory: While primarily read-only, the skill permits running "narrow, non-destructive checks" to verify claims (SKILL.md section 5).
  • Sanitization: There is no requirement for the agent to sanitize or filter the content of the files or PR descriptions it processes.
  • [COMMAND_EXECUTION]: The skill allows the agent to run "narrow, non-destructive checks" (SKILL.md Section 5) to resolve claims that cannot be determined from code alone.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 10:42 PM
Security Audit — agent-trust-hub — explain-change