OpenClaw Scheduler Token Auditor
v1.0.2Audit OpenClaw scheduler token usage for cron jobs, scheduled tasks, and heartbeat sessions. Use when the user wants to know which scheduled job is expensive...
⭐ 1· 92·1 current·1 all-time
by@yueeli
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 instructions: the SKILL.md describes using OpenClaw cron and session APIs (or the openclaw CLI fallback) to collect run records and usage.total_tokens. Nothing requested (no env vars, no external downloads) looks unrelated to auditing scheduler token usage.
Instruction Scope
Runtime instructions stay on-task: they tell the agent to call cron list/runs and session APIs, to prefer authoritative usage.total_tokens, and to treat heartbeats as bounded when exact usage is absent. There are no instructions to read arbitrary filesystem paths or unrelated environment variables, nor to exfiltrate data to third-party endpoints.
Install Mechanism
Instruction-only skill with no install spec and no code files — lowest-risk category. The only fallback is a platform CLI example (openclaw), which is reasonable and not itself an installer.
Credentials
No required env vars or credentials declared. The work described reasonably requires access to OpenClaw scheduler/session records; that is proportional to the auditing purpose. Users should be aware the agent executing the skill will need whatever platform access/permissions are normally required to call those APIs.
Persistence & Privilege
always is false and the skill does not request persistent system-wide changes. Autonomous invocation is allowed (default) but not combined with other red flags.
Assessment
This is an instruction-only skill that appears coherent and conservative: it relies on OpenClaw scheduler/session APIs (or the openclaw CLI) and explicitly requires authoritative fields like usage.total_tokens rather than estimating. Before enabling it, confirm that the agent identity you run it under already has read access to scheduler run records and session transcripts — those transcripts can contain sensitive prompt or user data. If you want tighter control, restrict the skill from autonomous invocation or run audits under an account with read-only, minimal-scope permissions. If you need assurance about the specific API calls the agent will make, request the implementation details or a list of concrete endpoints it will call.Like a lobster shell, security has layers — review code before you run it.
auditvk97et4j1zb40gmewwxfqkd3gzx83d1vwlatestvk97ds8s0mm59p0e90qezmngyc183hmysopenclawvk97et4j1zb40gmewwxfqkd3gzx83d1vw
License
MIT-0
Free to use, modify, and redistribute. No attribution required.