Skills
Navigate
skills/stayhydated/skills/manage-workspace-agents-md/Gen Agent Trust Hub

manage-workspace-agents-md

Pass

Audited by Gen Agent Trust Hub on Oct 13, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from repository files (such as AGENTS.md, READMEs, and manifests) which could contain malicious instructions designed to influence the agent's behavior during the maintenance workflow.
  • Ingestion points: The skill reads repository evidence, including existing AGENTS.md files, READMEs, manifests, and source code, as defined in SKILL.md and patterns/evidence-and-scope.md.
  • Boundary markers: The skill implements an "Explicit invocation gate" in SKILL.md to prevent unintended activation and uses "placeholder-hygiene" checks in patterns/evidence-and-scope.md to sanitize its output.
  • Capability inventory: The skill has the capability to write or patch AGENTS.md files within the local filesystem when the agent operates in Draft or Patch mode.
  • Sanitization: The instructions require the agent to classify repository claims (Observed vs. Inferred) and mandates the removal of template placeholders and unchecked links before finalizing any changes.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 13, 2026, 03:21 PM