F.A.C.E.T. Framework (心智模型：F.A.C.E.T. 框架)
v1.0.2F.A.C.E.T. cognitive framework for extracting mental models from classic books and theories. Use when (1) user asks for "reading notes", "extract models", or...
⭐ 0· 104·0 current·0 all-time
byKeDouPi(珂抖屁)@kedoupi
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name and description (F.A.C.E.T. mental-model extraction) match the declared behavior: no binaries, no credentials, no installs, and only an optional USER.md read for personalization. All required capabilities are proportional to the stated purpose.
Instruction Scope
SKILL.md instructs the agent to read USER.md (explicitly declared in the skill header) to personalize [T] Transfer. Apart from that, the runtime instructions are narrowly scoped to producing the F.A.C.E.T. analysis and the structured KB metadata block. Note: reading USER.md is appropriate for personalization but could expose whatever the user put in that file (see guidance).
Install Mechanism
No install spec and no code files — purely instruction-only. This minimizes risk because nothing is downloaded or written to disk by the skill itself.
Credentials
The skill requests no environment variables, no credentials, and no config paths beyond an optional USER.md. The only I/O is reading USER.md for personalization and producing a Markdown + YAML metadata block — proportionate for the described task.
Persistence & Privilege
always:false and default autonomous invocation are present. The skill does not request persistent presence or elevated privileges and does not modify other skills or system configs. It emits a metadata block intended for the caller to store, which is normal behavior for a content-generation skill.
Assessment
This skill appears internally consistent and low-risk: it only reads an optional USER.md to tailor the [T] Transfer and outputs a F.A.C.E.T. analysis plus a fenced metadata block. Before installing, (1) inspect USER.md and remove any secrets or sensitive personal data you would not want the agent to read, (2) confirm how the caller (or host agent) will store the KB metadata — the skill itself does not perform storage but outputs a block that may be saved elsewhere, and (3) ensure you trust the agent/caller that will invoke this skill (it could read USER.md whenever invoked). If you see unexpected network or filesystem behavior at runtime (e.g., data sent to external endpoints or writes to other files), revoke the skill and investigate — that would change the assessment.Like a lobster shell, security has layers — review code before you run it.
FACETvk970wvvm9zyzfwrr5hcz7wbdyx83jz9nbooksvk970wvvm9zyzfwrr5hcz7wbdyx83jz9ncognitive-toolsvk970wvvm9zyzfwrr5hcz7wbdyx83jz9ndecision-makingvk970wvvm9zyzfwrr5hcz7wbdyx83jz9nframeworksvk970wvvm9zyzfwrr5hcz7wbdyx83jz9nlatestvk97a0pysc8jd02e1fkhzmndcf183pebzlearningvk970wvvm9zyzfwrr5hcz7wbdyx83jz9nmental-modelsvk970wvvm9zyzfwrr5hcz7wbdyx83jz9npersonalizationvk970wvvm9zyzfwrr5hcz7wbdyx83jz9nstrategic-thinkingvk970wvvm9zyzfwrr5hcz7wbdyx83jz9n
License
MIT-0
Free to use, modify, and redistribute. No attribution required.