The Agent Skills Directory

[SAFE]: The skill operates as a set of informational templates and instructions for the agent to behave as a procurement professional. It does not include any third-party dependencies, network requests, or command executions.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-supplied procurement context to generate documents. This is a standard functional surface with no identified malicious triggers.
Ingestion points: User input gathered in SKILL.md (Step 1) and project details defined in templates/prompt-main.md.
Boundary markers: The instructions do not use specific delimiters for user-provided context but follow structured markdown templates.
Capability inventory: The skill instructions involve writing markdown output to files. No other tools or dangerous capabilities are requested or used.
Sanitization: The skill does not implement specific sanitization or filtering logic for processed user data.

procurement-pro