The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process potentially untrusted data from various external and project-level sources to generate specifications.
Ingestion points: As described in SKILL.md, the agent reconstructions context from "code, current docs, prior specs, tickets, and evidence," and fetches "current primary documentation" for external libraries, APIs, and cloud contracts.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore potential commands embedded within the analyzed content.
Capability inventory: The skill has the authority to write and update markdown files within the docs/specs/ directory and pass data to subsequent tools like to-tickets.
Sanitization: There are no explicit instructions for sanitizing or escaping content retrieved from external sources before interpolation.
[EXTERNAL_DOWNLOADS]: The skill instructions mandate network operations to retrieve information from external sources.
The Process section in SKILL.md directs the agent to "Fetch current primary documentation when an external library/API/CLI/cloud contract matters," which requires accessing external domains to pull technical details into the agent's context.

to-spec