ai-skill-integration-guide

Pass

Audited by Gen Agent Trust Hub on Jun 20, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill facilitates indirect prompt injection by instructing the agent to ingest and process data from untrusted external GitHub repositories.
  • Ingestion points: The agent is guided to read external repository README files, list all skill files, and extract frontmatter metadata (SKILL.md, line 35-45).
  • Boundary markers: No instructions or delimiters are provided to distinguish between the skill's trusted integration logic and the untrusted data being imported.
  • Capability inventory: The integration process involves shell command execution (find, sed, grep), execution of an internal audit script (audit-skills.sh), and the ability to update library documentation and configuration files (SKILL.md, line 51-54, 99-106).
  • Sanitization: The guide lacks instructions for sanitizing, validating, or escaping external content before it is interpolated into the agent's context or used to generate new skill files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 20, 2026, 09:57 PM
Security Audit — agent-trust-hub — ai-skill-integration-guide