The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill connects to an external MCP server at 'https://workorai.com/mcp' and references a remote npm package '@workorai/agent-kit'. These resources are part of the intended service infrastructure and are necessary for the skill's primary talent-marketplace functionality.
[PROMPT_INJECTION]: The skill contains an attack surface for indirect prompt injection because it processes untrusted external data.
Ingestion points: External data enters the context via employer.search_candidates_for_job, employer.get_candidate_evidence, and employer.get_applicant_transcript in SKILL.md.
Boundary markers: The skill does not explicitly define delimiters or instructions to ignore embedded commands within the processed candidate data.
Capability inventory: The agent has the ability to perform stateful marketplace actions such as candidate.apply_to_job and employer.invite_candidate across its scripts.
Sanitization: There is no mention of sanitization or filtering for the external content before it is processed by the agent.
[SAFE]: The skill implements best practices by explicitly instructing the agent to obtain user confirmation before executing stateful or visible marketplace actions (e.g., applying, inviting, or changing job statuses).

workorai