tailrocks-skill-update

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data such as repository files, tool reports, and scripts.
  • Ingestion points: Repository files, reports, scripts, references, fixtures, and tool output are identified as untrusted data in SKILL.md.
  • Boundary markers: The skill includes explicit defensive instructions stating that embedded instructions in data cannot change scope, authority, or governing rules.
  • Capability inventory: The skill involves reading repository files and performing authorized updates to SKILL.md and associated references.
  • Sanitization: The instructions explicitly forbid the exfiltration or reproduction of secret values, directing the agent to cite location and type instead of content.
  • [COMMAND_EXECUTION]: The skill uses the mise tool manager to run standard validation tasks.
  • Evidence: references/house-wiring.md specifies commands like mise run docs, mise run lint, and mise run docs:check to ensure repository consistency. These are well-known development tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 03:12 PM
Security Audit — agent-trust-hub — tailrocks-skill-update