ClawHub

公司内部系统PRD编写

v1.0.1

Use this skill when the user wants to design a brand-new (0→1) PRD or prototype for a company-internal process management system, where roles and workflows a...

⭐ 0· 52·0 current·0 all-time
byliu@lailiai
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 claim to produce PRD, workflows, RBAC, tracking plan and an HTML prototype — the SKILL.md and reference files describe exactly those outputs. There are no environment variables, binaries, or install steps that would be unrelated to producing documents and an HTML prototype.
ℹ
Instruction Scope
Instructions stay focused on product-design tasks (input collection, validations, user checkpoints, ER/flows/RBAC/tracking and a single-file HTML prototype). One notable instruction is domain-shift detection: Step 2.1 requires scanning the session/conversation history to detect field changes. This is relevant to the task but means the skill will read prior conversation context; users should be aware that previous messages may be used as input. The skill does not instruct exporting data to external endpoints, but the prototype spec uses Tailwind CDN (a public CDN) for styling — browser fetches to the CDN are expected when viewing the generated HTML.
✓
Install Mechanism
Instruction-only skill with no install spec and no code files; this minimizes on-disk installation risk. The prototype is produced as single HTML using Tailwind CDN and native JS (no packaged installers or downloads).
✓
Credentials
No environment variables, credentials, or config paths are requested. All required inputs are collected interactively from the user (roles, pain points, expectations, decisions). This is proportionate to the stated functionality.
✓
Persistence & Privilege
always:false and standard autonomous invocation apply. The skill does not request persistent presence or modify other skills or system-wide settings. It writes structured artifacts under a local project/ layout (as part of its workflow), which is expected for a document/prototype generator.
Assessment
This skill appears coherent and focused on producing internal PRDs and a single-file prototype, but consider these practical points before installing: - Confirm you're comfortable with the agent reading prior conversation history (Step 2.1) because it uses that context to detect domain shifts. - Avoid supplying production secrets or PII when authoring input; generated prototype HTML may include sample data and will reference Tailwind CDN (viewing the HTML causes a browser to fetch that CDN). - The skill intentionally refuses to invent API/interface details and punts technical design to a separate tech-design skill — if you need API specs, plan to run that downstream skill. - Review any generated prototype HTML before deploying or sharing; ensure it does not embed real credentials, analytics keys, or third‑party scripts beyond the stated Tailwind CDN. If you want higher assurance, request the author to confirm: (a) whether the agent will persist the project/ files to a shared location, and (b) whether any telemetry or analytics are enabled when the prototype is rendered.

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

latestvk975v9h1dbgpzsb42ybs0k9gdx84xnsr

License

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

Comments