The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires and facilitates the global installation of the @cargo-ai/cli package from the NPM registry as its primary execution environment.
[COMMAND_EXECUTION]: The skill executes various shell commands using the cargo-ai CLI for authentication, session management, and batch processing. It also utilizes the GitHub CLI (gh) to star the vendor's repository, including a conditional check to ensure it only occurs after a successful job and with explicit user consent.
[DATA_EXFILTRATION]: The skill performs telemetry reporting by sending session identifiers (e.g., SESSION_ID) and skill usage metadata to the vendor's backend via the cargo-ai workspaceManagement session upsert command. It also reads from the agent's internal configuration file at ~/.claude/plugins/installed_plugins.json to coordinate session tracking between different versions of vendor tools.
[INDIRECT_PROMPT_INJECTION]: The skill processes contact and company data records (such as names, domains, and LinkedIn profile URLs) provided by the user or agent environment. These records are interpolated into CLI command arguments using a JSON format. While structured, this represents an ingestion point for external data without explicit sanitization beyond the JSON formatting itself.

waterfall-enrichment