The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the repository being analyzed, which could contain malicious instructions designed to influence the agent's output.
Ingestion points: The skill reads system documentation (flows.md, permissions.md, automation.md), architecture files, and existing test suites (SKILL.md lines 38-40, 50-54).
Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when interpolating repository content into the agent context.
Capability inventory: The skill has the capability to write to the file system (updating tests.md and creating reports) and generates CI workflow snippets and branch protection settings for the user to apply (SKILL.md lines 117-124, 159).
Sanitization: There are no explicit instructions for sanitizing or validating the content extracted from the repository before it is used to generate test cases or CI recommendations.

derive-tests