lean-gate

Pass

Audited by Gen Agent Trust Hub on Nov 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to run "relevant deterministic checks" using the tools already present in the project's environment. While the instructions include a safety directive to "Inspect commands before execution," this capability necessitates the use of shell execution to run tests and validation scripts.
  • [INDIRECT_PROMPT_INJECTION]: The skill's workflow involves reading external data sources such as project instructions, manifests, and CI configurations to determine which checks are mandatory. These files constitute an untrusted data ingestion surface where malicious instructions could be embedded to manipulate the agent's verification logic.
  • Ingestion points: Project instructions, manifests, and CI configuration files.
  • Boundary markers: None explicitly defined for the ingested data.
  • Capability inventory: Execution of local project tools and commands as identified in Step 2.
  • Sanitization: No specific sanitization or escaping of the ingested file content is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 1, 2026, 11:18 AM