Skills
Navigate
skills/cadence-lab/artifact-cafe-skill/artifact-cafe/Gen Agent Trust Hub

artifact-cafe

Pass

Audited by Gen Agent Trust Hub on Nov 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from user-provided files (artifacts) and external feedback (comments), which could contain instructions intended to influence agent behavior.
  • Ingestion points: Local directory files during publish, remote artifact files downloaded via pull, and review threads fetched via comments and listen commands.
  • Boundary markers: Absent. The skill does not define structural markers for the ingested content, though the resulting HTML is sandboxed on the server-side.
  • Capability inventory: Network operations (HTTP/HTTPS via curl and npx), file system read/write access.
  • Sanitization: Includes a detailed "Content policy" section instructing the agent to evaluate and refuse the publication of malicious or sensitive content.
  • [COMMAND_EXECUTION]: Executes shell scripts (scripts/publish.sh, scripts/comments.sh) and invokes the artifact-cafe CLI tool using npx.
  • [EXTERNAL_DOWNLOADS]: Fetches the artifact-cafe package from the npm registry and interacts with the artifact.cafe API to upload and download artifact data.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 11, 2026, 08:16 AM