Metrics

Capture, normalize, and report metrics across any domain with reusable dimensions, programmable formulas, and scalable reporting workflows.

This skill appears coherent and low-risk because it is instruction-only, stores data locally under ~/metrics/, and requests no credentials. Before installing: 1) Confirm you understand and approve the 'activate whenever the user discusses KPIs...' setup option (decline if you don't want proactive monitoring/suggestions). 2) Expect the skill to create and update ~/metrics/ files — review those files for any sensitive content and do not store secrets there. 3) If you later install the related skills (analytics, dashboard, sql, etc.), review those skills separately because they may request external credentials or installs. 4) If you need stricter guarantees, test in a disposable account or environment and inspect the created ~/metrics/ files after first use.

Type: OpenClaw Skill Name: metrics Version: 1.0.0 The skill bundle is benign, consisting entirely of documentation and metadata without any executable code. Key indicators include explicit statements in `SKILL.md`, `memory-template.md`, and `setup.md` that the skill does not exfiltrate data, does not access files outside its designated `~/metrics/` directory, and explicitly instructs the agent to 'Never store secrets, credentials, or unrelated personal data'. The instructions for the AI agent are focused on responsible metric management and setup, with no evidence of prompt injection for malicious purposes.