Skip to content
Agentmart
Submit a tool

trace-debug

Read run traces and choose the next action when a fetch/extract fails.

Skill by ZenRows

IndexedMIT★ 16
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/zenrows-trace-debug

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Read run traces and choose the next action when a fetch/extract fails.

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/trace-debug

What's inside

Skill name
trace-debug
Description
Read run traces and choose the next action when a fetch/extract fails.
License
MIT
Instructions
77 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Trace & debug

Every `fetch`/`extract` writes a trace under `.zenrows/runs/<run-id>/` and
`.zenrows/traces/<run-id>/` (secret-free).

## Commands
```
zenrows trace inspect <run-id>    # raw run record
zenrows trace explain <run-id>    # what happened + likely cause + next action + exact command
zenrows trace replay <run-id>     # reconstruct the command line
zenrows trace export <run-id>     # JSON for sharing
```

## Failure → action map
- `FETCH_FAILED` / empty content → first retry in auto mode (`mode=auto`), which
  escalates for you and bills only for what succeeds. Only when auto mode has
  failed on its own, take manual control with `--manual --js-render`, then
  `--premium-proxy`. For slow pages, add `--wait-for <selector>` to a
  `--manual --js-render` run; auto mode may ignore it. Each step
  multiplies the cost of the request, and both together are the most expensive
  configuration available ([[cost-control]]).
- `REQUEST_TIMEOUT` → the CLI stopped waiting; the API was reached. Raise
  `--timeout` (default 120000ms, above the API's own 90s budget), or drop
  `--wait-for` so the request finishes inside that budget and the API returns
  its own error. Not a connectivity problem — do not chase the network.
- `BACKEND_UNAVAILABLE` → a genuine transport failure (DNS/TCP/TLS). Check
  connectivity and `zenrows config show`.
- `AUTH_INVALID` → re-check the key, `zenrows login --api-key …`.
- `KEY_CREDIT_CAP_REACHED` → this API key hit one of its own credit caps (HTTP
  402, gateway code `AUTH014`, Batch `api_key_cap_reached`). The account still
  has credits and its other keys keep working. Do not retry and do not escalate:
  `zenrows usage` shows the cap and when it resets. Wait for the reset, or have
  the account owner raise or remove the cap at
  https://app.zenrows.com/settings/api-keys.
- `POLICY_MAX_CREDITS_EXCEEDED` → the account itself is out of credits (or a
  local policy credit limit was hit). Do not retry-loop; `zenrows usage` shows
  when credits renew. Top up or upgrade, or claim the account if it is an
  auto-created Free plan.
- `BATCH_FAILED` → the batch run ended `failed` for a reason other than
  credits, subscription or key cap (those get `POLICY_MAX_CREDITS_EXCEEDED`,
  `AUTH_INVALID` or `KEY_CREDIT_CAP_REACHED`, with the run's `failure_reason`
  as `server_code` and its `job_id`); `failure_detail` in
  `zenrows batch status <id> --json` says why. Once fixed, run the unfinished
  tasks with `zenrows batch rerun <id> --status failed,pending,processing`;
  do not resubmit. Also a 400 (fix the flagged tasks) or a 5xx (retry).
- `BATCH_QUOTA_EXCEEDED` → 429: 3 active jobs (shared with Crawl), 5 scheduled
  jobs, or an open job's task cap. Nothing was created; free a slot (see
  `zenrows batch list` and `zenrows crawl list`).
- `BATCH_ACCESS_DENIED` / `BATCH_NOT_FOUND` → 403 Batch not enabled, or 404
  wrong or deleted job id. Do not retry.
- `BATCH_RUN_IN_PROGRESS`, `BATCH_RUN_ENDED`, `BATCH_NO_MATCHING_TASKS`,
  `BATCH_RERUN_WINDOW_EXPIRED`, `BATCH_REQUEST_IN_FLIGHT`, `BATCH_CONFLICT`,
  `BATCH_TASK_NOT_READY`, `BATCH_TASK_FAILED`, `BATCH_RESULT_EXPIRED` → the
  Batch API's 409/410/422 answers; `server_code` names the case and
  `next_action` the fix. See [[batch-jobs]].
- `CRAWL_NOT_ENABLED` → the API answered 403 `REQS008`: Crawl is not
  enabled for this account. Do not retry; ask Zenrows support for access.
- `CRAWL_INSUFFICIENT_CREDITS` / `CRAWL_KEY_CAP_REACHED` → Crawl's 402: the
  account is out of credits, or this key reached its credit cap (`AUTH014`).
  The same codes come back for a crawl that credits ended (`stopped`, or
  `failed` on older deployments; `server_code` `insufficient_credits` /
  `api_key_cap_reached`); its fetched pages stay readable.
  Same advice as `POLICY_MAX_CREDITS_EXCEEDED` / `KEY_CREDIT_CAP_REACHED`.
- `CRAWL_TOO_MANY_CRAWLS` → 429 `too_many_crawls`: the account has reached
  its limit of active jobs (3 by default), shared with its Batch jobs.
  Nothing was created; retry after `retry_after` seconds (see
  `zenrows crawl list`).
- `CRAWL_FAILED` → the crawl ended `failed`, or `stopped` with an error
  other than credits; `error.code` in
  `zenrows crawl get <id> --json` says why, with `error.detail`. See
  [[crawl]].
- `PARAM_CONFLICT_AUTO_MANUAL` → drop the managed flags or add `--manual`.
- `CAPABILITY_UNAVAILABLE` → the primitive is not available on this account (e.g. beta/invite-only); use the local-spec path where offered.

Escalate only with evidence from the trace. See [[cost-control]].

From ZenRows/cli, MIT.

More from ZenRows/cli

Signals

Stars
16
Forks
2
Open issues
0
Last commit (this folder)
Oct 20, 2026 (8 days ago)
Last release
Oct 23, 2026 (5 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from ZenRows

Cite this page

trace-debug skill. Agent Mart. https://agentmart.dev/skills/zenrows-trace-debug (last updated 2026-10-24)