pipeline-health-analyzer-skill
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data, creating a potential injection surface.
- Ingestion points: The skill explicitly requests "Pipeline export (CSV with deal data) or CRM access" as input in
SKILL.md. - Boundary markers: There are no instructions provided to the agent to treat the CSV content as untrusted or to use specific delimiters to prevent the data from being interpreted as instructions.
- Capability inventory: The skill includes a file-writing capability, instructing the agent to save an outcome sidecar to the local file system.
- Sanitization: No validation or sanitization steps are defined for the data processed from the CSV export.
- [COMMAND_EXECUTION]: The skill instructs the agent to perform an out-of-band file write operation.
- Evidence: The
Emit Outcome Sidecarsection inSKILL.mddirects the agent to write a JSON object to~/.claude/skill-analytics/last-outcome-pipeline-health-analyzer.json. While intended for telemetry, writing to hidden application directories in the home folder without explicit user prompts for each session is a behavior that requires oversight.
Audit Metadata