maple-python-style
Warn
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: MEDIUMCREDENTIALS_UNSAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [CREDENTIALS_UNSAFE]: The skill recommends inlining the MAPLE_KEY directly in the source code. The instructions explicitly advise against reading these keys from environment variables, which increases the risk of accidental credential exposure if the source code is committed to a repository.
- [INDIRECT_PROMPT_INJECTION]: The skill records exceptions directly into telemetry spans using span.record_exception(exc).
- Ingestion points: Raw exception data is processed in the error handling section of SKILL.md.
- Boundary markers: None provided; the skill does not suggest using delimiters to separate exception content from metadata.
- Capability inventory: Data is transmitted to external endpoints via OTLPSpanExporter and OTLPLogExporter.
- Sanitization: No sanitization or filtering is applied to the exception data before it is recorded and exported.
Audit Metadata