Skills
Navigate
skills/getcargohq/gtm-skills/find-linkedin-url/Gen Agent Trust Hub

find-linkedin-url

Pass

Audited by Gen Agent Trust Hub on Oct 5, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests profile data from LinkedIn via the Cargo API and passes the results to a second command for validation without sanitization.
  • Ingestion points: candidates.json (SKILL.md) containing results from findProfileUrl action.
  • Boundary markers: None.
  • Capability inventory: cargo-ai orchestration action execute-batch (SKILL.md) which executes remote actions and writes local files.
  • Sanitization: None; uses jq to interpolate data directly into command-line arguments.
  • [COMMAND_EXECUTION]: The skill executes several shell commands to manage the Cargo platform and interact with GitHub.
  • Evidence: npm install -g @cargo-ai/cli, cargo-ai login, cargo-ai workspaceManagement session upsert, and gh api -X PUT /user/starred/getcargohq/gtm-skills.
  • [DATA_EXPOSURE]: The skill reads the content of ~/.claude/plugins/installed_plugins.json to detect plugin conflicts and decide whether to track the session.
  • Evidence: grep -q '"cargo@gtm"' ~/.claude/plugins/installed_plugins.json in SKILL.md.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 5, 2026, 08:45 PM