The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill uses the import_job_by_url tool to fetch and process content from external job boards and arbitrary URLs.
Ingestion points: The skill ingests untrusted data from platforms like LinkedIn, Greenhouse, Lever, and Workday, or any URL provided to import_job_by_url (referenced in SKILL.md).
Boundary markers: Absent. The instructions do not provide delimiters or specific warnings for the agent to ignore instructions embedded within the job descriptions it imports.
Capability inventory: The skill possesses high-impact capabilities including apply_to_job, send_outreach, update_profile, and upload_resume as described in SKILL.md.
Sanitization: Absent. There is no evidence of content sanitization or validation before the imported data is processed by the AI.
[EXTERNAL_DOWNLOADS]: The skill setup requires the installation of an external package and connection to a remote server.
Evidence: The setup instructions in SKILL.md direct users to install jobgpt-mcp-server via npx and connect to a remote endpoint at https://mcp.6figr.com/mcp.

jobgpt