ln-97-observability-engineer
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves ingesting and analyzing untrusted project artifacts such as source code, manifests, and lockfiles. This establishes a surface where malicious instructions embedded in a project's files could influence the agent's behavior.
- Ingestion points: Reads project instructions, manifests, lockfiles, and configuration files (SKILL.md).
- Boundary markers: Uses a structured 'Execution contract' and 'Output contract' to define behavior and reporting boundaries.
- Capability inventory: Includes file reading and scoped editing capabilities via project-native tools.
- Sanitization: Includes explicit mandates to redact secrets and sensitive payloads, and to sanitize untrusted fields before processing data (Checklist item 2).
Audit Metadata