ClawHub

51mee Resume Profile

v1.2.1

简历画像。触发场景：用户要求生成候选人画像；用户想了解候选人的多维度标签和能力评估。

⭐ 0· 251·0 current·0 all-time
by51mee@51mee-com
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name/description (resume profiling) aligns with the instructions: read an uploaded resume, extract text, send to a large model, and return structured JSON. No unrelated environment variables, binaries, or install steps are requested.
✓
Instruction Scope
SKILL.md stays within scope: it instructs reading user-uploaded resumes, extracting text, calling a model with a strict prompt/JSON schema, and returning results. It does not instruct reading other files, system paths, or extraneous environment variables. The prompt explicitly forbids fabricating fields and asks to ignore prompt-tampering attempts (injection protection).
✓
Install Mechanism
Instruction-only skill with no install spec and no code files — nothing is written to disk or downloaded by the skill itself, which minimizes install risk.
✓
Credentials
No credentials, env vars, or config paths are required. The requested access (processing uploaded resumes) is proportional to the stated purpose; there are no unrelated secret requests.
✓
Persistence & Privilege
Skill is not always-enabled and makes no persistent system-level changes in its instructions. It does not request modifying other skills or global agent config.
Assessment
This skill appears coherent for resume analysis, but it will send extracted resume text to whatever large-model endpoint the agent uses — resumes commonly contain sensitive personal data (PII). Before installing, confirm: (1) which model/endpoint will be used and its data retention/privacy policy; (2) that your environment has or permits the necessary PDF/DOC/JPG->text extraction (OCR) if you expect images; (3) you accept that candidate PII (names, contact info, ID numbers) may be transmitted to the model provider; and (4) you test with non-sensitive samples first. If you need stricter privacy, require on-premise/in-house model hosting or redact sensitive fields before analysis.

Like a lobster shell, security has layers — review code before you run it.

latestvk97bgq06ve32yfnjyyzyecr909831qd8

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Comments