plan-antislop

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is designed as a planning and auditing tool. It explicitly prohibits automated rewrites or code modifications, requiring user approval for each phase of its generated burndown plan.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it audits untrusted codebase content and user-facing copy. However, the risk is mitigated by its limited capabilities: it only generates a markdown report (plan-antislop.md) and does not execute code, perform network requests, or modify existing files based on the audited data.
  • Ingestion points: Processes files, strings, and UI components within a provided codebase (SKILL.md).
  • Boundary markers: Instructions require the agent to quote only the minimum necessary content to identify "tells" rather than processing entire files, and strictly limits actions to planning only.
  • Capability inventory: No subprocess calls, no exec/eval, no network operations, and no writing to source files; only emits a planning document.
  • Sanitization: Employs minimal quoting and a structured report template to prevent control flow issues.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 05:56 AM
Security Audit — agent-trust-hub — plan-antislop