skills/backnotprop/pstack/interrogate/Gen Agent Trust Hub

interrogate

Pass

Audited by Gen Agent Trust Hub on Jul 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the 'git diff' command to identify changes for review. This is a legitimate use of tool capabilities within the context of a code review assistant.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface when processing untrusted code for review by subagents. (1) Ingestion points: Code diffs, file contents, commit messages, and PR descriptions are gathered in SKILL.md. (2) Boundary markers: The {DIFF_OR_FILES} placeholder in references/reviewer-prompt.md lacks delimiters or encapsulation to separate untrusted code content from reviewer instructions. (3) Capability inventory: The skill uses the Task tool to spawn multiple subagents. (4) Sanitization: There is no evidence of sanitization or escaping of the ingested code content before it is interpolated into subagent prompts.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 4, 2026, 05:53 PM
Security Audit — agent-trust-hub — interrogate