The Agent Skills Directory

[PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it is designed to process and analyze untrusted content from user resumes.
Ingestion points: The skill accepts resumes in PDF, Word, Markdown, or plain text formats for analysis as described in the "Step 1: 接收简历" section of SKILL.md.
Boundary markers: Absent. The instructions do not provide the agent with specific delimiters (e.g., triple backticks) or "ignore embedded instructions" warnings for handling the user's resume content.
Capability inventory: The skill's capabilities are restricted to text analysis and generating a markdown report. It does not perform any file system writes, network requests, or subprocess executions.
Sanitization: Absent. There are no instructions for sanitizing or escaping the content within the resume before processing.
[NO_CODE]: The skill consists solely of instructions in SKILL.md and does not include any scripts, binaries, or executable logic, which significantly reduces the overall risk profile.

resume-ats