remove-ai-slops

Pass

Audited by Gen Agent Trust Hub on Jul 19, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes source code from branch diffs that may contain malicious instructions.
  • Ingestion points: Source code content and file lists are ingested through git diff and file reads in Phases 1 and 4.
  • Boundary markers: The skill lacks explicit delimiters to separate untrusted code from instructions.
  • Capability inventory: The agent can modify files using the Edit tool, execute shell commands through project runners, and orchestrate sub-agents.
  • Sanitization: No sanitization is performed on the ingested code before analysis.
  • [COMMAND_EXECUTION]: The skill performs local command execution and manages dynamic script execution.
  • It uses git, awk, and wc for repository analysis and file metrics.
  • It executes local project tools for testing, linting, and type-checking.
  • It generates and runs regression tests in Phase 2 to verify that cleanup operations do not alter existing program behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 19, 2026, 05:10 AM
Security Audit — agent-trust-hub — remove-ai-slops