coingecko-open-page

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a well-defined workflow for navigating to historical data pages on a reputable external service (CoinGecko).
  • [COMMAND_EXECUTION]: The skill executes a local Python script (scripts/build-url.py) using the uv tool. This script serves as a security control by validating the coin-id against a strict regular expression (^[a-z0-9]+(?:-[a-z0-9]+)*$) and applying URL encoding before constructing the final destination URL.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it processes output from the cg search tool. However, this is mitigated by the following controls:
  • Ingestion points: Data enters via the output of cg search in SKILL.md step 1.
  • Boundary markers: None explicitly defined in the prompt instructions.
  • Capability inventory: The skill can open new pages in Chromium via mcp__chrome-devtools__new_page.
  • Sanitization: The build-url.py script performs rigorous regex validation and character escaping on the ingested data before it is used to form a URL, preventing path traversal or URL manipulation attacks.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 04:27 PM
Security Audit — agent-trust-hub — coingecko-open-page