narrative-registry

Pass

Audited by Gen Agent Trust Hub on Jul 5, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's instructions and metadata are free from malicious patterns, obfuscation, or unauthorized capability requests.
  • [EXTERNAL_DOWNLOADS]: The skill references the author's official GitHub repository as its project homepage.
  • [PROMPT_INJECTION]: The skill identifies and mitigates a potential surface for indirect prompt injection arising from the processing of external brand content.
  • Ingestion points: The skill ingests user-provided text and candidate data from the memory/narrative-registry/ directory.
  • Boundary markers: The instructions include explicit directives to treat all external material as untrusted data rather than instructions, referencing the project's security protocol.
  • Capability inventory: Capabilities are restricted to writing text to specific local files (e.g., canon.md, versions.md, candidates.md). No network access or code execution tools are requested.
  • Sanitization: The logic explicitly prohibits untrusted data from influencing administrative registry functions, such as version control or term approval lists.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 5, 2026, 04:16 PM
Security Audit — agent-trust-hub — narrative-registry