boss-agent-cli
Pass
Audited by Gen Agent Trust Hub on Jun 17, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the 'boss-agent-cli' package and 'chromium' via the 'uv' and 'patchright' package managers to provide its core functionality. These resources are provided by the skill's author or established technology providers.
- [EXTERNAL_DOWNLOADS]: Contains instructions in 'evals/README.md' to download the official evaluation harness from the Anthropic GitHub repository for performance measurement.
- [PROMPT_INJECTION]: This skill possesses an indirect prompt injection surface as it processes external job listing data from web platforms. This risk is addressed via several mechanisms:
- Ingestion points: Untrusted data enters the context through job search results and detail queries defined in 'SKILL.md'.
- Boundary markers: Present; 'SKILL.md' explicitly instructs the agent to only parse structured JSON envelopes from stdout and treat stderr strictly as logs.
- Capability inventory: The skill utilizes the 'boss' CLI to perform searches, view details, manage a local shortlist, and export results.
- Sanitization: Present; the skill enforces a strict JSON output contract ('schema_version 1.0') to ensure data is handled in a structured format rather than free-form text.
Audit Metadata