strk20-privacy

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill interacts with several external services and sources:
  • The utility script scripts/check_freshness.py performs network requests to the NPM registry (registry.npmjs.org) and GitHub (raw.githubusercontent.com) to verify the latest versions of Starknet-related packages.
  • The script performs liveness checks on documentation pages at strk20-by-example.org.
  • The documentation provides instructions to add a sibling skill via npx skills add starkience/strk20-agent-skills.
  • [COMMAND_EXECUTION]: The SKILL.md instructions recommend executing a bundled Python script, scripts/check_freshness.py, which uses the Python standard library to perform connectivity and version checks.
  • [INDIRECT_PROMPT_INJECTION]: The skill acts as a router for external documentation, creating an inherent surface for indirect prompt injection if the source material were to be compromised.
  • Ingestion points: Local markdown files in the references/ directory and external documentation fetched by the agent from strk20-by-example.org.
  • Boundary markers: Not explicitly defined within the instruction set.
  • Capability inventory: The skill allows for local script execution and the installation of additional agent skills.
  • Sanitization: No specific sanitization or filtering of the ingested documentation content is described.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 10:11 PM
Security Audit — agent-trust-hub — strk20-privacy