agent-scout-explorer
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill implements a workflow that ingests untrusted external data, creating a potential surface for indirect prompt injection.
- Ingestion Points: Scans arbitrary codebase files, dependencies, logs, and external dependency definitions (such as
package.jsonandREADME.md). - Boundary Markers: Absent; there are no structural boundaries or instructions configured to prevent the agent from executing malicious commands embedded within scanned files.
- Capability Inventory: Leverages the
mcp__claude-flow__memory_usagetool to write data fields directly into the swarm coordination space. - Sanitization: No data validation or input filtering is performed prior to parsing and passing contents into the storage memory structure.
Audit Metadata