setup-work-skills

Pass

Audited by Gen Agent Trust Hub on Aug 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a bundled shell script (scripts/install.sh) to perform installation tasks, including directory management and symlink creation.
  • [EXTERNAL_DOWNLOADS]: The installer fetches additional skill components by cloning the author's official GitHub repository (https://github.com/kenlck/skills.git).
  • [INDIRECT_PROMPT_INJECTION]: The do-work and do-seo-work skills ingest data from GitHub issue bodies and PRDs, presenting a standard surface for indirect prompt injection.
  • Ingestion points: Processes external data via gh CLI commands reading issue descriptions and comments in files like do-work/SKILL.md and do-seo-work/SKILL.md.
  • Boundary markers: No specific delimiters or "ignore instructions" warnings are defined for the ingested issue content.
  • Capability inventory: The skills possess capabilities to write to the file system, create Git branches, and perform authenticated GitHub API mutations using the gh tool.
  • Sanitization: The instructions do not specify sanitization or validation of the retrieved issue text before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 18, 2026, 02:25 AM
Security Audit — agent-trust-hub — setup-work-skills