Intent Translation Skill

v1.0.0

引导用户表达模糊想法,挖掘真实需求,生成结构化指令,帮助AI精准理解和执行用户意图。

0· 136·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
medium confidence
Purpose & Capability
Name/description (translate vague user intent into structured prompts) aligns with the included scenario files and the translate.py logic. Required resources are minimal (no env vars, no binaries) and match the stated purpose.
Instruction Scope
SKILL.md instructs a confined dialog flow (ask 3–4 questions, build a prompt, return it). It does not request reading unrelated system files, credentials, or sending data to external endpoints in the prose. The scenarios and templates only reference user answers and local templates.
Install Mechanism
No install spec provided (instruction-only plus a local script). No downloads, package installs, or external installers are declared.
Credentials
No environment variables, credentials, or config paths are required. The scenario data and script operate on bundled JSON files and in-memory answers only.
Persistence & Privilege
always is false and the skill is user-invocable; it does not request permanent or elevated presence. Nothing in SKILL.md or visible files attempts to modify other skills or system-wide configuration.
Assessment
This skill appears coherent and only uses local scenario templates plus a translator script to build prompts. Before installing or running it: (1) inspect the full scripts/translate.py end-to-end (the displayed file was truncated) to confirm there are no network calls (requests, urllib, sockets), subprocess.exec usage, or os.environ accesses that could exfiltrate data; (2) avoid entering secrets (passwords, API keys, private keys, bank details) into the dialog because answers become part of generated prompts and could be copied elsewhere; (3) run the skill in a restricted environment if you want extra safety (e.g., no network) until you verify the code; (4) if you plan to use generated prompts with other AI services, be aware those services will receive whatever user-provided content you include. If you share the complete translate.py file tail here, I can re-check for any hidden network or I/O behavior to raise confidence to high.

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

latestvk972fmfxe6sa19gwh2mzv70fqn83az8n

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments