colosseum-copilot

Pass

Audited by Gen Agent Trust Hub on Aug 1, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill makes network requests to https://copilot.colosseum.com to retrieve project data, winner patterns, and research archives.
  • [COMMAND_EXECUTION]: Uses local shell commands and inline Python snippets to manage its configuration file (~/.superstack/config.json), log telemetry events, and perform API calls via curl.
  • [CREDENTIALS_UNSAFE]: Asks the user for a Personal Access Token (PAT) to authenticate with the Colosseum API. The token is stored locally in the user's configuration directory, which is a standard method for managing API credentials in developer tools.
  • [DATA_EXFILTRATION]: Transmits basic telemetry data (skill name, execution phase, status, and platform architecture) to a backend service defined in the local configuration. The telemetry implementation includes an opt-out mechanism and does not transmit PII, source code, or the API token.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 1, 2026, 11:04 AM
Security Audit — agent-trust-hub — colosseum-copilot