AI Adoption Audit
v1.0.0企业 AI 落地情况调查工具。输入公司名+简介，输出 AI 成熟度体检报告：推动层级、部署方式、模型选型、解决场景、推广范围、ROI、问题与教训。支持市场调研和客户参考两种输出模式。Triggers: AI落地调查, 企业AI, ai adoption, company AI audit, AI成熟度, 公司AI...
⭐ 0· 54·0 current·0 all-time
byKen Wang@kenwang-dev
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name, description and runtime instructions are coherent: the skill is a structured, web-research-based auditor that gathers public sources and produces a formatted report. It does not request unrelated credentials, binaries, or installs.
Instruction Scope
SKILL.md instructs multi-round web searches, cross-checking sources, and producing detailed markdown reports — all consistent with the purpose. It also explicitly instructs the agent to check and use a local archive (skills/ai-adoption-audit/output/) and to read previous reports to perform incremental updates; this file I/O and caching behavior is not declared elsewhere and could cause stale responses or persist previously generated report contents.
Install Mechanism
No install spec and no code files — instruction-only skill. Lowest install risk (nothing will be downloaded or executed beyond the agent following the written instructions).
Credentials
The skill declares no environment variables, credentials, or config paths. Its required actions (web search, compile public data, write reports) do not inherently need secrets, so lack of credential requests is proportional.
Persistence & Privilege
The skill expects persistent local storage (a specific output directory and history snapshots) and logic to return cached results when last-run ≤ 7 days. It does not set always:true or request elevated privileges, but the implicit requirement for file-system read/write access should be confirmed before enabling (and prior reports could contain sensitive content).
Assessment
This skill appears to do what it says: structured public-web research and report generation, and it asks for no credentials or installs. Before installing or enabling it, verify: (1) whether your agent environment will actually allow it to read/write the stated path (skills/ai-adoption-audit/output/) and where those files are stored, (2) retention and access policies for generated reports (they may contain sensitive findings), (3) whether cached reports will be returned automatically (the skill will avoid re-searching if a report is ≤ 7 days old), and (4) whether you prefer the skill to always perform fresh searches. If you have privacy or compliance concerns, test the skill in a sandboxed agent, ask the author to declare or make configurable the archive path and cache/retention policy, and avoid running it on machines that contain sensitive internal documents you don't want included in archives.Like a lobster shell, security has layers — review code before you run it.
latestvk9758z5wy9bkz23tzx9pnnaptn84akrs
License
MIT-0
Free to use, modify, and redistribute. No attribution required.