linktree-ci-integration

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process repository specifications and workflow files, which represents a potential untrusted data surface.
  • Ingestion points: Reads repository specifications, workflow files, and campaign fixtures (SKILL.md).
  • Boundary markers: The instructions explicitly direct the agent to report "redacted diagnostics" and use "sanitized fixtures" to mitigate risks in untrusted pull requests.
  • Capability inventory: The skill uses Write and Edit tools to modify workflow files and tests based on processed data.
  • Sanitization: The skill mandates synthetic tests that do not contact external services and requires the use of canonical producers for generated output to prevent drift.
  • [EXTERNAL_DOWNLOADS]: The skill uses WebFetch to access official Linktree documentation and status pages. These references target a well-known service and are used for informational purposes only.
  • [REMOTE_CODE_EXECUTION]: The instructions explicitly prohibit fetching and executing remote code via WebFetch, restricting its use to verifying public behavior or approved documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 02:51 AM
Security Audit — agent-trust-hub — linktree-ci-integration