模型消耗统计
v1.0.0模型消耗统计技能，用于统计和展示不同会话、不同模型的 token 使用量、成本统计和消耗报告，通过飞书消息卡片进行可视化展示。当用户请求查看模型消耗、token 统计、成本分析或消耗报告时使用。
⭐ 0· 47·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description match the runtime instructions: the SKILL.md only describes collecting session usage (via session_status, sessions_list, sessions_history), computing token/cost metrics, and sending formatted Feishu cards. No unrelated credentials, binaries, or installs are requested.
Instruction Scope
Instructions are scoped to reading session data and producing reports sent through the platform 'message' tool (Feishu cards). This is appropriate, but it assumes the 'session_*' tools supply accurate token details and that the 'message' tool is authorized to post to Feishu; verify those tool permissions and what session data they expose (conversation contents may be included).
Install Mechanism
No install spec and no code files — instruction-only skill. Nothing will be downloaded or written to disk by this skill itself.
Credentials
The skill declares no environment variables, credentials, or config paths. That is proportionate for a reporting skill that relies on platform-provided session and messaging tools.
Persistence & Privilege
always is false and the skill is user-invocable (normal). It does not request persistent/enforced presence or modify other skills. Note: the platform default allows autonomous invocation, but that is not unique to this skill and always:false reduces continuous privilege.
Assessment
This skill appears internally consistent and low-risk, but verify two operational details before enabling: (1) Confirm what data the platform 'session_status'/'sessions_list' APIs expose (they may include full conversation text); if they include sensitive content, restrict who can invoke the skill or filter/redact fields before reporting. (2) Confirm how the 'message' tool posts to Feishu (which Feishu account/app is used) and that no unintended data is shared externally. Also check that the model pricing numbers match your billing rates if you rely on the cost calculations. If you want extra safety, allow this skill only for manual/user-invoked reports (disable autonomous invocation where possible) and review generated card previews before sending to external channels.Like a lobster shell, security has layers — review code before you run it.
latestvk97by8kkc03eeec8yktrbs8zz184rm4b
License
MIT-0
Free to use, modify, and redistribute. No attribution required.