strava-api
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The scripts 'refresh_token.py', 'setup_oauth.py', and 'strava_api.py' all attempt to install the 'requests' library from the Python Package Index at runtime if it is not present in the environment.
- [COMMAND_EXECUTION]: The skill executes shell commands using 'os.system' for package management and 'subprocess.run' to interact with system-level credential managers (macOS Keychain and Linux Secret Service) for secure token storage.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the Strava API that could potentially contain malicious instructions. 1. Ingestion points: Athlete profiles, activity descriptions, and activity comments are fetched via the StravaClient in 'scripts/strava_api.py'. 2. Boundary markers: The skill does not implement explicit delimiters or instruction-ignore warnings when presenting API data to the agent. 3. Capability inventory: The skill has the ability to execute network requests (GET/POST/PUT/DELETE), write to the local file '~/.strava/metadata.json', and interact with system keychains. 4. Sanitization: No specific sanitization or filtering of external API content is performed before the data is processed.
- [DYNAMIC_EXECUTION]: The skill uses 'os.system' to dynamically invoke the Python interpreter for pip installations during execution.
Audit Metadata