triangulate-diagnosis
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from user-provided symptoms, error logs, and repo content, which could theoretically contain malicious instructions aimed at influencing the agent's behavior.
- Ingestion points:
SKILL.md(Phase 0) describes gathering exact symptoms, error text, and environment notes from the user to create a shared brief for subagents. - Boundary markers: While the skill enforces a
secret-redactionboundary for evidence entering the brief, it does not explicitly define delimiters or "ignore instructions" markers for the user-supplied bug brief text itself. - Capability inventory: The skill possesses the ability to write to the local filesystem (saving specs and tickets to
docs/specs/anddocs/tickets/) and orchestrate multiple sub-skills (diagnose,to-spec,to-tickets,wayfinder). - Sanitization: The instructions explicitly include a pass for secret redaction to prevent credential exposure but do not specify sanitization or escaping of potential prompt injection payloads within error logs or symptoms.
Audit Metadata