Official agent skill

Cron Watchdog Debug

by vercel-labs in vercel-labs/vercel-openclaw-archived

Cron and watchdog debugging for vercel-openclaw: Vercel Cron auth, persisted OpenClaw jobs, cron wake keys, token refresh, restore oracle, hot spare, and watchdog reports.

OfficialMITAuto-check passedProductivity & Automation

Install Cron Watchdog Debug

skills CLI
$ npx skills add vercel-labs/vercel-openclaw-archived --skill cron-watchdog-debug -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install vercel-labs/vercel-openclaw-archived cron-watchdog-debug --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/vercel-labs/vercel-openclaw-archived.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/cron-watchdog-debug .claude/skills/cron-watchdog-debug && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
cron-watchdog-debug
GitHub stars
117
Token cost
~1.6k tokens
SKILL.md length
548 words
Files
2 (incl. references)
Skills in repo
16
Repo updated
First seen
Licence
MIT

At a glance

Cron and watchdog debugging for vercel-openclaw: Vercel Cron auth, persisted OpenClaw jobs, cron wake keys, token refresh, restore oracle, hot spare, and watchdog reports.

  • Works in 4 steps: Deployment-state proof. → Watchdog path diagram with every edge… → Hypothesis table with fastest falsifier… → …
  • Scheduled OpenClaw jobs fail to wake
  • SKILL.md covers Non-Negotiables, Evidence First, Watchdog Path Diagram and Runtime Probes, plus 5 more sections
  • Calls node, git and npx

What it does

Cron Watchdog Debug is an agent skill from vercel-labs/vercel-openclaw-archived, published by the product's own GitHub organization. Cron and watchdog debugging for vercel-openclaw: Vercel Cron auth, persisted OpenClaw jobs, cron wake keys, token refresh, restore oracle, hot spare, and watchdog reports. Use when scheduled OpenClaw jobs fail to wake or run, watchdog status is wrong, cron persistence is suspect, or /api/cron/watchdog behavior changes.

Its SKILL.md is about 1.6k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files, including reference files (for example `references/handoff-template.md`).

It sits in Productivity & Automation, covering Scheduled and recurring tasks. It works with Vercel. The repository describes itself as: Deploy OpenClaw on Vercel. The licence is MIT.

When your agent uses it

  • Scheduled OpenClaw jobs fail to wake
  • Watchdog status is wrong
  • Cron persistence is suspect
  • /api/cron/watchdog behavior changes

Example prompts

  • “/cron-watchdog-debug”

Requirements

  • Node.js

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

  1. Deployment-state proof.
  2. Watchdog path diagram with every edge marked unknown, verified-good, or verified-bad.
  3. Hypothesis table with fastest falsifier and status.
  4. Cron Watchdog Handoff.

What it can do on your machine

Read from SKILL.md and the folder at commit d592f7b. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • node
    • git
    • npx

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use git and npx, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Cron Watchdog Debug loads about 1.6k tokens when it runs, and up to ~2k if it reads all its reference files. Until then it costs about 85 tokens; SKILL.md has 548 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~85
When it runs · the whole SKILL.md, loaded when a task matches
~1.6k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~2k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from vercel-labs/vercel-openclaw-archived at commit d592f7b, republished under its MIT licence (© vercel-labs). 548 words, ~1,602 tokens.

Download SKILL.mdSave it as .claude/skills/cron-watchdog-debug/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
cron-watchdog-debug
description
Cron and watchdog debugging for vercel-openclaw: Vercel Cron auth, persisted OpenClaw jobs, cron wake keys, token refresh, restore oracle, hot spare, and watchdog reports. Use when scheduled OpenClaw jobs fail to wake or run, watchdog status is wrong, cron persistence is suspect, or /api/cron/watchdog behavior changes.

Cron Watchdog Debug

Use this skill for cron wake, watchdog, scheduled OpenClaw job, and restore-oracle incidents.

Non-Negotiables

Before proposing a fix, produce:

  1. Deployment-state proof.
  2. Watchdog path diagram with every edge marked unknown, verified-good, or verified-bad.
  3. Hypothesis table with fastest falsifier and status.
  4. Cron Watchdog Handoff.

Do not treat "cron route ran" as proof that an OpenClaw scheduled job ran. Keep these states separate: Vercel Cron invoked, watchdog authorized, persisted wake due, sandbox woke, AI Gateway token refreshed, OpenClaw cron scheduler loaded jobs, user-visible delivery happened.

Evidence First

For live deployments, create one artifact root:

bash
RUN_TS="$(date -u +%Y%m%dT%H%M%SZ)"
ART=".agent-runs/cron-debug/$RUN_TS"
mkdir -p "$ART"/{admin,vercel,sandbox,workflow}

Collect before editing code:

  • .vercel/project.json versus the intended project/team/deployment.
  • git rev-parse HEAD and git ls-remote origin main.
  • Live GET /api/status.
  • Live GET /api/admin/sandbox-diag.
  • Live GET /api/admin/logs filtered for watchdog., sandbox., gateway., cron, and restore.
  • Live GET /api/admin/launch-verify if the incident involves launch readiness.
  • Live GET or POST /api/cron/watchdog only with valid cron authorization.

If a release env file is supplied, source it only in the shell running live commands. Never print or save raw secrets.

Watchdog Path Diagram

Use this shape and mark each edge:

text
Vercel Cron tick -> /api/cron/watchdog auth -> runSandboxWatchdog
  -> deployment contract -> metadata read -> busy/stale/running/stopped branch
  -> cron wake key read -> ensureSandboxReady? -> token refresh
  -> OpenClaw cron jobs present in sandbox -> scheduled job executes -> delivery visible

For a running sandbox due soon, include the parallel branch:

text
running sandbox -> gateway probe -> cronNextWakeMs <= now + 60s
  -> force AI Gateway token refresh -> restore oracle cycle -> no duplicate wake

Runtime Probes

Read admin surfaces first. Then, only if needed, inspect the sandbox with npx sandbox or the app admin SSH/exec fallback.

Read-only sandbox checks:

bash
set -eu
echo "== process =="
ps -eo pid,ppid,comm,args | grep -E "[o]penclaw|[n]ode" || true
echo "== ports =="
(ss -ltnp || netstat -ltnp) 2>/dev/null | grep -E ":(3000|8787)\\b" || true
echo "== cron files =="
ls -la /home/vercel-sandbox/.openclaw/cron 2>/dev/null || true
echo "== cron jobs shape =="
node - <<'NODE'
const fs = require('fs');
for (const p of [
  '/home/vercel-sandbox/.openclaw/cron/jobs.json',
  '/home/vercel-sandbox/.openclaw/cron/jobs-state.json',
]) {
  try {
    const j = JSON.parse(fs.readFileSync(p, 'utf8'));
    console.log(p, JSON.stringify({
      version: j.version ?? null,
      jobCount: Array.isArray(j.jobs) ? j.jobs.length : Object.keys(j.jobs || {}).length,
      jobIds: Array.isArray(j.jobs) ? j.jobs.map((x) => x.id) : Object.keys(j.jobs || {}),
    }));
  } catch (err) {
    console.log(p, 'unreadable', err.message);
  }
}
NODE

Do not print job payload text if it may contain user data. Prefer shape, counts, ids, and next-run timestamps.

Store Evidence

Cron wake depends on store keys written through the lifecycle layer. Do not hardcode Redis prefixes in app code. For debugging, use app/admin surfaces when available; if direct store inspection is unavoidable, record key names and redacted shapes only.

High-signal facts:

  • cronNextWakeMs exists and is due, future, missing, or malformed.
  • cronJobsJson structured record exists, has a SHA-256, job count, and source.
  • lastRestoreMetrics.cronRestoreOutcome is present or missing after wake.
  • watchdog report contains cron.wake, token.refresh, probe, and restore.prepare checks.
Show full SKILL.md (232 more words)Show less

Hypotheses To Compare

Keep all rows visible as they are ruled out:

HypothesisEvidence forEvidence againstFastest falsifierStatus
Vercel Cron did not invoke the routeVercel logs for /api/cron/watchdog
Cron auth rejected the requestroute response/log status 401 vs report
Wake key was never persistedstore/admin evidence for cronNextWakeMs
Wake key is future or stalecompare cronNextWakeMs to current UTC
Watchdog intentionally skipped idle sandboxcron.wake check message and metadata status
Sandbox woke but token refresh failedtoken.refresh check and AI Gateway token metadata
Jobs file missing or malformed in sandboxsanitized cron file shape from sandbox
OpenClaw scheduler loaded jobs but delivery failedOpenClaw logs plus channel lastForward/user-visible evidence
Restore oracle or hot spare side effect changed cron staterestore.prepare check and restore metrics

Safe Fix Boundaries

  • src/app/api/cron/watchdog/route.ts owns cron-route auth and response shape.
  • src/server/watchdog/run.ts owns watchdog decision order and report checks.
  • src/shared/watchdog.ts owns report types.
  • src/server/sandbox/lifecycle.ts owns cron persistence keys, stop/touch persistence, and wake behavior.
  • src/server/sandbox/cron-persistence.test.ts and src/server/watchdog/run.test.ts own regression coverage.
  • lat.md/sandbox-lifecycle.md owns the architecture narrative for cron wake.

Do not edit channel routes while debugging cron unless evidence proves the scheduled job reached channel delivery and failed there. Switch to the relevant channel skill for that phase.

Verification

Pick the smallest verification that covers the changed edge:

bash
node scripts/verify.mjs --steps=test,typecheck
node --test src/server/watchdog/run.test.ts
node --test src/server/sandbox/cron-persistence.test.ts
lat check

For live fixes, include before/after evidence from /api/cron/watchdog, /api/admin/logs, and sandbox cron file shape.

Handoff

Use references/handoff-template.md for incident reports.

© vercel-labs, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file (references) in .agents/skills/cron-watchdog-debug of vercel-labs/vercel-openclaw-archived.

  • SKILL.md
  • references/handoff-template.md

Open the folder on GitHubat commit d592f7b

Compare with similar skills

Cron Watchdog Debug next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Cron Watchdog Debug compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Cron Watchdog Debug this skillvercel-labs/vercel-openclaw-archived117—~1.6kAutomated safety check: PassMIT
Cron Opsczl9707/build-your-own-openclaw1.9k—~593Automated safety check: PassMIT
Wp Wpcli And OpsAutomattic/agent-skills2112 repos~988Automated safety check: PassNone
Superset Automatesuperset-sh/superset15k—~1.4kAutomated safety check: PassCustom licence
Action Items Todoistmgonto/executive-assistant-skills118—~3.8kAutomated safety check: NotesNone
Symfony Schedulerdev-toolings/superpowers-symfony223—~526Automated safety check: NotesMIT

Similar skills

  • Cron Ops

    czl9707/build-your-own-openclaw

    Create, list, and delete scheduled cron jobs. An agent skill from czl9707/build-your-own-openclaw.

    1.9k GitHub stars~593 tokensUpdated 3 mo ago
    Productivity & AutomationAuto-check passed
  • Wp Wpcli And Ops

    Automattic/agent-skills

    A skill your agent uses when working with WP-CLI (wp) for WordPress operations: safe search-replace, db export/import, plugin/theme/user/content management, cron, cache flushing, multisite, and…

    211 GitHub starsUsed in 2 repos~988 tokens
    Productivity & AutomationAuto-check passed
  • Superset Automate

    superset-sh/superset

    Turns a recurring chore into a scheduled or event-triggered Superset agent, drafting its prompt, picking a target and trigger, and reviewing the first run.

    15k GitHub stars~1.4k tokensUpdated today
    Productivity & AutomationAuto-check passed
  • Action Items Todoist

    mgonto/executive-assistant-skills

    Extract action items from today's Granola/Grain meetings, create Todoist tasks, complete fulfilled tasks, and draft meeting-triggered follow-up emails.

    118 GitHub stars~3.8k tokensUpdated 7 mo ago
    Productivity & AutomationAuto-check: notes
  • Symfony Scheduler

    dev-toolings/superpowers-symfony

    Schedule recurring tasks with the Symfony Scheduler component (native since 7.x); define schedules, triggers, and integrate with Messenger

    223 GitHub stars~526 tokensUpdated 4 days ago
    Productivity & AutomationAuto-check: notes
  • Alibabacloud Ecs Windows Os Troubleshooting

    aliyun/alibabacloud-ecs-troubleshoot-skills

    Troubleshoot and repair Alibaba Cloud ECS Windows instances from inside the GuestOS or remotely via Cloud Assistant.

    148 GitHub stars~6.1k tokensUpdated 1 mo ago
    Productivity & AutomationAuto-check passed

More from vercel-labs/vercel-openclaw-archived

All 16 skills in this repo
  • Channel Debug Core

    vercel-labs/vercel-openclaw-archived

    Official

    Channel webhook triage for vercel-openclaw Slack/Telegram/Discord/WhatsApp issues: prove deployment state, collect admin readiness endpoints, build evidence-first handoff before fixes.

    117 GitHub stars~2k tokensUpdated 4 mo ago
    Auto-check: notes
  • Lat Md

    vercel-labs/vercel-openclaw-archived

    Official

    Writing and maintaining lat.md documentation files — structured markdown that describes a project's architecture, design decisions, and test specs.

    117 GitHub starsUsed in 1 repo~1.2k tokens
    Auto-check passed
  • Auth Store Debug

    vercel-labs/vercel-openclaw-archived

    Official

    Auth and store debugging for vercel-openclaw: admin-secret mode, Sign in with Vercel, session cookies, CSRF, LOCALREADONLY, Redis vs memory store, keyspace namespacing, and metadata shape migrations.

    117 GitHub stars~465 tokensUpdated 4 mo ago
    Auto-check passed
  • Firewall AI Gateway Debug

    vercel-labs/vercel-openclaw-archived

    Official

    Firewall and Vercel AI Gateway debugging for vercel-openclaw: network policy allowlists, OIDC token refresh, AI Gateway transform rules, firewall learning/enforcement, and sandbox.update…

    117 GitHub stars~550 tokensUpdated 4 mo ago
    Auto-check passed
  • Gateway Proxy Debug

    vercel-labs/vercel-openclaw-archived

    Official

    Gateway and proxy debugging for vercel-openclaw: /gateway routing, HTML injection, WebSocket rewrite, gateway-token handoff, waiting page, status heartbeat, sandbox port URL cache, and proxy auth.

    117 GitHub stars~573 tokensUpdated 4 mo ago
    Auto-check passed
  • Openclaw Bootstrap Debug

    vercel-labs/vercel-openclaw-archived

    Official

    OpenClaw bootstrap, bundle, config, and restore-asset debugging for vercel-openclaw: openclaw.bundle sidecars, plugin discovery, channel catalog, restart scripts, config hashes, dynamic resume…

    117 GitHub stars~494 tokensUpdated 4 mo ago
    Auto-check passed

Works with

Questions about Cron Watchdog Debug

What does Cron Watchdog Debug do?

Cron and watchdog debugging for vercel-openclaw: Vercel Cron auth, persisted OpenClaw jobs, cron wake keys, token refresh, restore oracle, hot spare, and watchdog reports. Cron Watchdog Debug is an agent skill from vercel-labs/vercel-openclaw-archived, published by the product's own GitHub organization. Cron and watchdog debugging for vercel-openclaw: Vercel Cron auth, persisted OpenClaw jobs, cron wake keys, token refresh, restore oracle, hot spare, and watchdog reports.

When should I use Cron Watchdog Debug?

Cron Watchdog Debug fits situations like: scheduled OpenClaw jobs fail to wake; watchdog status is wrong; cron persistence is suspect; /api/cron/watchdog behavior changes.

How do I install Cron Watchdog Debug in Claude Code?

Run `npx skills add vercel-labs/vercel-openclaw-archived --skill cron-watchdog-debug -a claude-code`. Or copy the skill folder (.agents/skills/cron-watchdog-debug in vercel-labs/vercel-openclaw-archived) into .claude/skills/cron-watchdog-debug in your project. Claude Code loads it when a task matches its description.

How do I install Cron Watchdog Debug in Codex?

Run `npx skills add vercel-labs/vercel-openclaw-archived --skill cron-watchdog-debug -a codex`. Or copy the skill folder (.agents/skills/cron-watchdog-debug in vercel-labs/vercel-openclaw-archived) into .agents/skills/cron-watchdog-debug in your project. Codex loads it when a task matches its description.

Can I use Cron Watchdog Debug in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add vercel-labs/vercel-openclaw-archived --skill cron-watchdog-debug -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/cron-watchdog-debug, .gemini/skills/cron-watchdog-debug, .github/skills/cron-watchdog-debug and .opencode/skills/cron-watchdog-debug in your project.

What does Cron Watchdog Debug need to run?

Going by SKILL.md and its folder, Cron Watchdog Debug needs the command-line tools its instructions call (node, git and npx). Our summary lists: Node.js.

Does Cron Watchdog Debug access the network?

SKILL.md contains no URLs. Its commands use git and npx, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Cron Watchdog Debug safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Cron Watchdog Debug use?

Cron Watchdog Debug is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Cron Watchdog Debug use?

About 1.6k tokens (SKILL.md is roughly 6.4k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 441 tokens, read only when the agent opens those files.

What are the alternatives to Cron Watchdog Debug?

Skills that share tags, products or a category with Cron Watchdog Debug: Cron Ops (czl9707/build-your-own-openclaw, 1.9k stars), Wp Wpcli And Ops (Automattic/agent-skills, 211 stars), Superset Automate (superset-sh/superset, 15k stars) and Action Items Todoist (mgonto/executive-assistant-skills, 118 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Cron Watchdog Debug?

vercel-labs (a GitHub organization, an official publisher) maintains it in vercel-labs/vercel-openclaw-archived, which has 117 GitHub stars. The repository holds 16 skills in this directory. The repository was last updated on May 15, 2026.

Source: vercel-labs/vercel-openclaw-archived on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.