Skillv1.0.1

ClawScan security

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

BenignMar 16, 2026, 9:24 AM

Verdict: Benign
Confidence: high
Model: gpt-5-mini
Summary: This is an instruction-only Didi (滴滴出行) assistant whose description, instructions, and code samples align with its stated purpose and request no unexpected credentials or installs.
Guidance: This skill is instruction-only and appears coherent with its stated purpose. It provides code samples that, if you run them, will call Didi enterprise APIs and require client_id/client_secret values — the skill itself does not ask for or store credentials. Before using: (1) do not paste production client_id/client_secret or other secrets into the chat; use test or least-privilege credentials when experimenting; (2) verify the API endpoints and company policy if integrating enterprise accounts; (3) if you plan to run the sample code, run it in a controlled/dev environment and inspect requests/responses; (4) prefer creating scoped/test API keys and rotate them after use. If you want, I can scan the rest of the truncated SKILL.md (or search for any hidden instructions) to raise confidence further.

Purpose & Capability: okThe name/description promise ride-planning, cost estimates, enterprise car management, and Open Platform API guidance; the SKILL.md contains product comparisons, cost-estimation code, high‑level strategies, and example enterprise-API client code — all consistent with that purpose.
Instruction Scope: okRuntime instructions are prose and example code for cost estimates and API usage. They reference the Didi enterprise API endpoint (api.es.xiaojukeji.com) which is appropriate for an API-integration guide. The instructions do not direct the agent to read unrelated system files, environment variables, or exfiltrate data.
Install Mechanism: okNo install spec and no code files — this is instruction-only, so nothing will be written to disk or downloaded by the skill itself.
Credentials: noteThe skill declares no required env vars or credentials. Example code shows use of client_id/client_secret and access tokens (expected for API integration). Because the skill doesn't request secrets automatically, there is no disproportionate credential access, but users should avoid pasting production credentials into chat.
Persistence & Privilege: okalways is false and the skill does not request persistent agent-level privileges or change other skills' configs. disable-model-invocation is default false (normal) and acceptable given no other red flags.