write-agent
Pass
Audited by Gen Agent Trust Hub on Feb 20, 2026
Risk Level: SAFE
Full Analysis
- SAFE (SAFE): The skill primarily consists of educational markdown files and templates. These files guide the creation of secure and efficient agent prompts without introducing any executable vulnerabilities.
- COMMAND_EXECUTION (SAFE): The script
scripts/validate-agent.shis a utility for linting local agent files. It uses standard Unix tools likegrep,sed, andawkto verify YAML frontmatter and markdown structure. It does not interact with the network or access sensitive system paths. - PROMPT_INJECTION (SAFE): The documentation includes sections on preventing 'designer failure modes' and avoiding 'escape-hatch' language in prompts. This promotes better security and reliability in the agents created using these templates.
- DATA_EXFILTRATION (SAFE): There are no network-capable tools, hardcoded credentials, or instructions that would lead to the exfiltration of sensitive data.
Audit Metadata