find-companies-using-tech

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the @cargo-ai/cli Node.js package globally. This is a vendor-owned tool required to interact with the Cargo API.
  • [COMMAND_EXECUTION]: The skill executes various cargo-ai CLI commands for authentication, workspace management, and data retrieval. It also utilizes the gh API to star the vendor's repository upon explicit user consent.
  • [DATA_EXFILTRATION]: The skill reads the ~/.claude/plugins/installed_plugins.json file to detect existing plugin installations. This information is used in a session upsert command sent to the vendor's API for attribution and telemetry purposes.
  • [PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection. 1. Ingestion points: User-provided technology names and job titles are interpolated into the --data JSON payload in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: The cargo-ai CLI executes remote actions and performs network operations. 4. Sanitization: Absent. The impact is limited to the data parameters sent to the vendor's orchestration API.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 10:07 AM
Security Audit — agent-trust-hub — find-companies-using-tech