strava
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes user-generated content from an external API without sanitization or explicit instruction boundaries, creating a surface for indirect prompt injection.\n
- Ingestion points: Data enters the agent's context through the output of
scripts/strava.pycommands, specificallylist_activities,get_activity_performance, andactivity, which return text fields like activity names, descriptions, and segment names.\n - Boundary markers: The
SKILL.mdinstructions do not define delimiters or provide specific instructions for the agent to disregard potentially malicious content embedded within the Strava data.\n - Capability inventory: The skill possesses capabilities to read activity details, stats, and streams. It does not have write access to the filesystem (except for its own credential management) or the ability to execute arbitrary shell commands.\n
- Sanitization: There is no evidence of sanitization, escaping, or filtering of the content retrieved from the Strava API before it is presented to the language model for summarization.
Audit Metadata