afk-coding-tradeoffs

Warn

Audited by Gen Agent Trust Hub on May 15, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to prompt the user to install an external dependency from a personal, non-authoritative GitHub repository if a specific tool is missing.
  • Evidence: In SKILL.md, the agent is directed to ask the user to run npx skills add https://github.com/leoreisdias/truss-framework if truss-evaluation is not found.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted data from various external project materials to inform its reasoning and output.
  • Ingestion points: The workflow in SKILL.md (Step 2: Load prior context) specifies reading roadmaps, planning documents, requirement notes, specs, and tickets.
  • Boundary markers: No explicit boundary markers or "ignore embedded instructions" warnings are defined for the ingested data.
  • Capability inventory: The skill has the capability to generate mandatory decision artifacts (files) based on the processed information.
  • Sanitization: There are no instructions for sanitizing or validating the contents of the external documents before they are incorporated into the agent's context.
Audit Metadata
Risk Level
MEDIUM
Analyzed
May 15, 2026, 09:17 PM
Security Audit — agent-trust-hub — afk-coding-tradeoffs