The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-supplied documentation and has the capability to write the resulting requirements to the repository.
Ingestion points: SKILL.md (Steps 1 and 2) instructions describe the intake of client briefs, feature requests, and meeting notes.
Boundary markers: The skill does not define specific delimiters or instructions to ignore potential commands embedded in the user-provided data.
Capability inventory: SKILL.md Step 6 ("Capture the deliverable") instructs the agent to write files to the repository.
Sanitization: No sanitization or validation logic is defined to check external data for malicious instructions before processing.
[SAFE]: The skill references technical standards and literature from well-known organizations including IEEE, INCOSE, and Springer for methodology guidance in references/source-literature.md.

krt-requirements-weaver