Repository navigation
fix(extract): describe extract=auto as it works today - #44
Merged
Merged
Conversation
The agent-facing `extract` tool description and the README still called extract=auto an "open beta" that falls back to autoparse "when the domain isn't enabled", which reads like a per-account gate. There is no gate: Extract works on any domain Zenrows has prepared, with any API key, and a domain not prepared yet answers 403 REQS007, which the tool already falls back from. Say that, and mark AUTH010 as the legacy code. Add extract, structured-data and web-data to the npm keywords so the package is findable for Extract. Copy and metadata only; no behaviour changes. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Extract is generally available. The README and the extract tool description agents read still labelled it beta. Drop the label and keep the note that it is currently free. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
AnderRV
approved these changes
Oct 13, 2026
When Extract knows a kind of page but none of the fields come back filled, the API answers 422 with code no_data_extracted. The request is not charged, and repeating it returns the same answer. The extract tool returned that as a generic "Zenrows error 422", which an agent reads as a failure to reach the site and tends to retry. It now returns ok: true with empty: true, data: null, noDataExtracted: true and a short message on how to check the page rendered fully. There is still one request and no Autoparse fallback. Other 422s, and a 422 outside extract=auto, keep returning an error. The tool description explains the new field. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
# Conflicts: # README.md
tedi-z
added a commit
that referenced
this pull request
Oct 14, 2026
#49 landed on main with its own version of the Batch-is-GA wording that 5aa2b50 added here, and #44 changed the README row next to it. These three files now carry main's wording for those lines, so the PR merges cleanly and its diff against main is the tool-wording change alone. Refs ACT-1566 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BUChpDttrHDtU8izeDT2op
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
extracttool description (what agents read):mode=autois no longer labelled beta (Extract is generally available; it stays currently free) and is described as working on any domain Zenrows has prepared with any API key, and falling back to autoparse when the domain isn't prepared yet (REQS007). Previously it said it falls back "when the domain isn't enabled".fallback_autoparseparam description and theisExtractUnavailablecomment:REQS007first,AUTH010marked as the legacy domain-not-enabled code.extract.extract,structured-data,web-data.Why
Public surfaces must match the current state of Extract (sibling PRs ZenRows/cli#36 and ZenRows/zenrows-python-sdk#40). Since the gate was removed, Extract works on any prepared domain with any active key; an unprepared domain answers
403 REQS007and can be prepared viaPOST /v1/extract/prepared-domains(https://docs.zenrows.com/extract/setup#domain-coverage).Copy and metadata only: no behaviour changes. Versions in
package.jsonandserver.jsonare untouched.Testing
npm run typecheckcleannpm run lintcleannpm test: 171/171 passAfter merge
Publish a new version so the npm page, the MCP registry entry and the hosted server pick up the new description and keywords.
🤖 Generated with Claude Code