Skillv1.0.0

ClawScan security

Dingding · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignMar 16, 2026, 9:17 AM

Verdict: Benign
Confidence: medium
Model: gpt-5-mini
Summary: Instruction-only DingTalk (钉钉) integration helper whose instructions and requirements are coherent with its stated purpose, but the package metadata lacks an external source/homepage which lowers provenance confidence.
Guidance: This skill is an instruction-only DingTalk API guide and appears coherent with that purpose. Before installing or using it: (1) recognize it has no code/install step — it only guides API usage, so it won’t itself store or use your secrets unless you paste them into the agent or an integration you build; (2) you will need to supply AppKey/AppSecret, webhook secret, or AES key to actually call or validate callbacks — keep these in a secure place (environment variables or secret manager), rotate them regularly, and never paste them into untrusted chat; (3) verify the agent/skill publisher and prefer skills with an identifiable source or homepage — this skill’s metadata lacks an external homepage, which reduces provenance confidence; (4) confirm all API endpoints are the official DingTalk domains (api.dingtalk.com / oapi.dingtalk.com) before sending requests; and (5) test on a non-production account first and restrict tokens to minimum scopes. Overall the skill is coherent and not requesting disproportionate access, but exercise normal caution with credentials and provenance.

Purpose & Capability: okThe SKILL.md content is focused on DingTalk/Open Platform APIs (tokens, webhooks, approvals, calendar, contacts, attendance, event callbacks). Nothing in the instructions requests unrelated services or credentials.
Instruction Scope: okRuntime instructions are limited to describing API endpoints, authentication flows, signing/encryption examples, and best practices. The skill does not instruct reading unrelated local files, calling non-DingTalk endpoints, or exfiltrating data.
Install Mechanism: okNo install spec and no code files are present (instruction-only), so nothing is written to disk or downloaded by the skill itself.
Credentials: noteThe skill declares no required environment variables or credentials (primaryEnv none). SKILL.md sensibly advises using AppKey/AppSecret, webhook secrets, and env vars to store them, but the skill won't automatically request those — you'll need to provide them when using the APIs. This is coherent but worth noting so users don't assume secrets are pre-configured.
Persistence & Privilege: okalways is false and it is user-invocable. The skill does not request persistent/privileged presence or make changes to other skills or system settings.