The Agent Skills Directory

[PROMPT_INJECTION]: The skill demonstrates an indirect prompt injection surface by processing untrusted data from external job descriptions.\n
Ingestion points: Untrusted data enters the agent context via job offers, links, or pasted descriptions as specified in SKILL.md, references/guide-tailor-cv.md, and references/guide-generate-cover-letter.md.\n
Boundary markers: There are no explicit delimiters or instructions provided to the agent to disregard embedded commands within the job description text.\n
Capability inventory: Across SKILL.md and referenced guides, the skill reads Candidate-Profile.md (which likely contains PII), creates new files in the local directory, and invokes external tools like the Researcher Agent and the track-application command/skill.\n
Sanitization: The skill does not perform any escaping or validation on the job description content before using it to generate tailored documents.

create-application