maple-agent-tracing-claude-agent-sdk
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEDATA_EXFILTRATIONCREDENTIALS_UNSAFE
Full Analysis
- [DATA_EXFILTRATION]: The skill provides instructions to configure the Claude Code CLI and Claude Agent SDK to export tracing information, including user prompts, tool arguments, and tool results, to the vendor's ingestion endpoints (ingest.maple.dev and ingest.eu.maple.dev). The documentation explicitly advises users to consider privacy and compliance constraints (such as PII or HIPAA) before enabling full content logging.- [CREDENTIALS_UNSAFE]: Instructions describe setting up an ingestion key using an Authorization header. The skill recommends using environment variables for secret management and provides code examples using placeholders and sentinel values for the actual keys.
Audit Metadata