gate-plan

Pass

Audited by Gen Agent Trust Hub on Jun 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes internal validation scripts located within the platform's plugin environment, specifically ${CLAUDE_PLUGIN_ROOT}/skills/gate-review/scripts/check-citations.py and ${CLAUDE_PLUGIN_ROOT}/skills/gate-review/scripts/check-inheritance.py. These tools are used to perform deterministic checks on the generated markdown files before they are presented to the operator.
  • [PROMPT_INJECTION]: The skill processes external, potentially untrusted data to generate validation criteria, creating an indirect prompt injection surface.
  • Ingestion points: Reads project-level files including QUEUE.md, Q{n}-why.md, and project README files (Step 1).
  • Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the ingested files.
  • Capability inventory: The skill has the capability to write files to the workspace (Q{n}-gate.md) and execute internal validation scripts.
  • Sanitization: No explicit sanitization or filtering of the ingested project data is performed before it is used to author the gate document.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 26, 2026, 04:30 AM
Security Audit — agent-trust-hub — gate-plan