finding-chain-correlator
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data, specifically
finding_setandapplication_context, creating a surface for potential injection attacks embedded within security reports. - Ingestion points: Data enters the agent's context through the defined inputs (
finding_set,application_context,auth_model,data_sensitivity_map) and is processed during Phase 1 (Normalization) and Phase 2 (Link Construction). - Boundary markers: The instructions lack specific guidance on using delimiters or strict boundary markers when interpolating these external inputs into the agent's reasoning process.
- Capability inventory: The provided skill content is purely instructional and does not include scripts or tools that perform subprocess calls, file system writes, or network operations.
- Sanitization: The workflow does not include explicit steps for the agent to sanitize or filter the content of incoming security findings for potentially malicious instructions.
Audit Metadata