Diagnose Scheduled Job Trigger Vs Execution Failure
v1.0.0用于排查“定时任务没执行”这类问题，并区分到底是未触发、已触发但执行失败，还是运行环境/授权失效导致的假象。遇到 cron 异常、任务未跑、自动任务失灵、网关重启后要验证恢复、怀疑是模型导致任务失败、需要查看日志作证、需要给出证据口径、要确认 `deactivated_workspace` / OAuth tok...
⭐ 0· 83·0 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
The name and description match the SKILL.md: the skill is explicitly about diagnosing scheduled-job trigger vs execution failures and its steps focus on logs, run records, error codes (deactivated_workspace, oauth token invalid, timeout) and recovery verification. Nothing requested (no env vars, no installs) is unrelated to that goal.
Instruction Scope
The SKILL.md instructs the agent to inspect run histories, logs, configuration files, and to manually trigger/verify representative jobs. Those actions are appropriate for this diagnostic purpose, but they necessarily require read (and sometimes write/trigger) access to system logs, run records and config paths. The instructions are procedural rather than prescriptive about exact commands; that gives an agent broad discretion to access system state, which is expected for troubleshooting but worth bounding in production (least privilege, audit).
Install Mechanism
Instruction-only skill with no install spec and no code files. This minimizes supply-chain risk (nothing is downloaded or installed).
Credentials
The skill declares no required environment variables, credentials, or config paths. The content discusses OAuth tokens and workspace states as diagnostic artefacts but does not request secrets or unrelated cloud credentials. That is proportionate to its stated investigative purpose.
Persistence & Privilege
always: false and no installation changes are requested. The skill does not ask to persist credentials or modify other skills. Autonomous invocation is allowed by platform default but is not requested to be forced (no always:true).
Scan Findings in Context
[no_code_files_instruction_only] expected: The scanner found no code files and only SKILL.md + evals.json. For a diagnostic/operational guide, being instruction-only is expected and reduces install-time risk.
[evals_contain_log_paths_and_prompts] expected: evals/evals.json contains realistic test prompts and example file paths (/srv/openclaw/openclaw.json, /opt/agent/openclaw.json, /var/log/openclaw/gateway.log). This is expected for an ops diagnostic skill, but those prompts demonstrate the skill intends to read system logs/configs—ensure the agent is permitted to access those paths before using in production.
[pre_scan_injection_signals_none] expected: No pre-scan injection signals were detected. This is consistent with an instruction-only skill that contains no embedded remote endpoints or obfuscated code.
Assessment
This skill is coherent for diagnosing scheduled-job problems: it tells an agent to check scheduler run records, logs, error codes and to re-run representative jobs to verify recovery. That requires the agent (or the person following the steps) to have access to logs/configs and the ability to trigger jobs — so before installing or enabling it, confirm the agent will only be granted the minimal, audited access needed (read access to the specific log directories and run records, and controlled ability to trigger test runs). Be cautious about running the skill in production without approvals: do not grant broad filesystem or cloud credentials just because the guide references OAuth tokens or openclaw.json. Finally, review the included evals.json for any environment-specific or sensitive paths and, if you plan autonomous invocation, ensure organizational policy allows an automated agent to read those artifacts and perform test triggers.Like a lobster shell, security has layers — review code before you run it.
latestvk979taqzm48vvj540bgv5f5xhs83dav9
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🕒 Clawdis