The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Instained the @cargo-ai/cli package from the official NPM registry to facilitate interaction with the vendor's job detection API.
[COMMAND_EXECUTION]: Executes shell commands including npm for installation, the cargo-ai CLI for business logic, and grep for environment discovery within the agent's plugin configuration.
[DATA_EXFILTRATION]: Performs a session management call to the vendor's infrastructure via cargo-ai workspaceManagement session upsert to provide attribution and tracking for the skill's usage.
[DYNAMIC_EXECUTION]: Suggests the use of npx skills add to dynamically integrate and load additional agent skills from the vendor's repository.
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: Processes contact data such as professional emails and domains as input for batch execution.
Boundary markers: No specific delimiters or safety instructions are used to isolate untrusted data.
Capability inventory: The skill can execute shell commands, perform network operations, and read local configuration files.
Sanitization: Data is interpolated into a JSON payload for the CLI, providing basic structural isolation, but no explicit sanitization logic for the values is present.
[SAFE]: Reads the file ~/.claude/plugins/installed_plugins.json to detect if the primary Cargo GTM plugin is already installed, ensuring idempotent behavior and preventing redundant session attribution.

track-job-changes