Query The Workflow Finder's 385+ tool and 206+ workflow catalog directly and get structured JSON back. Built for AI agents, integrations, and anyone who wants the data instead of the page.
No API key, no signup, and no cost. The endpoint reads the same live data every tool and workflow page on this site renders from, so results never drift out of sync with what a human visitor sees.
Send a GET request with your query as URL parameters:
curl "https://www.theworkflowfinder.com/api/v1/search?task=automate+social+media+posts&limit=1"Response:
{
"query": {
"task": "automate social media posts",
"limit": 1
},
"tools": [
{
"slug": "sprout-social",
"name": "Sprout Social",
"url": "https://www.theworkflowfinder.com/tools/sprout-social",
"category": "Marketing",
"pricing": "paid",
"pricingDetail": "$199/seat/mo Standard / $299/seat/mo Professional / $399/seat/mo Advanced (30-day trial, no free tier)",
"shortDescription": "Run publishing, sentiment-aware inbox triage, and competitor listening for a whole social team from one enterprise dashboard.",
"rating": 4.4,
"reviewCount": 4300,
"review": {
"status": "not-reviewed",
"qualityScore": null,
"limitedScore": null,
"scale": 5,
"scope": null,
"testedAt": null,
"publishedAt": null,
"reviewUrl": null,
"reason": "No published, evidence-backed product-quality review yet.",
"methodologyUrl": "/review-methodology"
},
"fitScores": [
{
"useCase": "Larger marketing and social teams needing a unified inbox with sentiment-aware prioritization for high message volume",
"capability": 5,
"reliability": 4,
"value": 3,
"ease": 3,
"compatibility": 4,
"overall": 3.8,
"rationale": "Directly the tool's bestFor and standout: the Smart Inbox routes every comment and DM into one queue with sentiment analysis flagging priority, matching pros on enterprise-grade reporting and cutting through high message volume."
},
{
"useCase": "A small company or solo social media manager mainly scheduling posts",
"capability": 2,
"reliability": 3,
"value": 1,
"ease": 2,
"compatibility": 2,
"overall": 2,
"rationale": "The tool's own ourTake and con state the per-seat pricing with no free tier is impossible to justify for smaller companies relative to lighter alternatives, and the full feature set requires the top Advanced plan."
}
],
"externalRating": null,
"legacyRating": {
"value": 4.4,
"reviewCount": 4300,
"provenance": "unverified",
"deprecated": true,
"notice": "Legacy rating/reviewCount fields are not TWF Review Scores or verified on-site review totals."
},
"bestFor": "Larger marketing and social teams that need a unified inbox with sentiment-aware prioritization, not just scheduling",
"ourTake": "Sprout Social is enterprise social media management, and per-seat pricing at $199 to $399 per month with no free tier signals who it's built for. The Smart Inbox routes every comment, mention, and DM into one queue and uses sentiment analysis to flag which messages need a human response first, a real operational advantage for teams handling high message volume. Smaller companies and solo social managers will find the price impossible to justify relative to lighter alternatives. The full feature set requires the top Advanced plan. Evaluate it when you're managing a large team with real message volume, not when scheduling posts is the main job.",
"capabilities": [
"Social media content",
"Ad campaign management"
],
"facts": {
"privacyPolicyUrl": "https://sproutsocial.com/legal/privacy-policy/",
"trustPageUrl": "https://sproutsocial.com/trust-center/",
"openSource": false,
"apiAvailability": "public",
"pricingVerifiedAt": "2026-08"
},
"score": 4,
"matchedOn": [
"4 keyword matches"
]
}
],
"workflows": [
{
"slug": "automate-social-media-posts",
"title": "Automate Social Media Posts",
"url": "https://www.theworkflowfinder.com/workflows/automate-social-media-posts",
"difficulty": "Intermediate",
"timeEstimate": "1 hr setup / 5 min weekly",
"description": "Learn how to automatically generate and schedule social media posts from your blog articles using AI and automation tools.",
"expectedResult": "A fully automated pipeline that drafts and schedules tweets and LinkedIn posts every time you publish a new blog post.",
"tags": [
"Social Media",
"Automation",
"Marketing"
],
"score": 4,
"matchedOn": [
"4 keyword matches"
]
}
],
"meta": {
"totalToolMatches": 129,
"totalWorkflowMatches": 96,
"source": "https://www.theworkflowfinder.com",
"generated": "2026-08-28",
"scoreContractVersion": "1.0",
"scoring": {
"methodologyUrl": "https://www.theworkflowfinder.com/review-methodology",
"review": {
"statuses": [
"not-reviewed",
"limited",
"reviewed",
"needs-update",
"blocked"
],
"scale": 5,
"qualityScore": "Product-quality Review Score only when status is reviewed; otherwise null, never zero.",
"limitedScore": "A narrower published editorial rating, not a product-quality Review Score or a quality-ranking input.",
"nullMeaning": "No publishable product-quality score; not a negative quality judgment. See status, reason and scope."
},
"fitScores": {
"methodologyUrl": "https://www.theworkflowfinder.com/fit-score",
"scale": 5,
"meaning": "Existing use-case-specific editorial Fit Scores, not product-quality Review Scores or measured benchmarks. An empty array means no Fit Scores are recorded."
},
"externalRating": "Attributed third-party reference with publisher, sourceUrl and checkedAt; null when no verified reference is available. Not a TWF Review Score.",
"legacyRating": {
"deprecatedFields": [
"rating",
"reviewCount"
],
"notice": "Legacy numeric fields are preserved unchanged where exposed. Use legacyRating for their provenance and deprecation notice; they are not TWF Review Scores or verified on-site review totals."
},
"searchScore": "score is query relevance, not a Review Score or a use-case Fit Score.",
"ranking": "Query relevance first; ties and empty queries use eligible product-quality reviews, then curated featured status and alphabetical name/slug. Unreviewed and limited tools have no numeric quality rank; legacy ratings and review counts are never ranking inputs."
}
}
}Score contract version 1.0 is an additive extension, not a replacement for existing fields. Tool results from this API, MCP search/lookup/compare tools, and /tools.json expose the same review, fitScores, externalRating, and legacyRating fields. Search metadata carries scoreContractVersion and scoring; the bulk dataset carries these at the top level. Examples on this page are generated from the current catalog, including these additive fields.
review.qualityScore: an evidence-backed product-quality Review Score out of 5, available only with status: reviewed. No qualifying evidence means null, never 0 and never an implied negative judgment. Read reason, scope, testedAt, publishedAt, and reviewUrl alongside it; unavailable context is also null.review.limitedScore: an existing narrower editorial rating out of 5, only applicable to the published review scope. A Limited review is not a comprehensive product-quality score and is not eligible for quality ranking. Read the Review Score methodology.fitScores: the unchanged array of use-case-specific Fit Scores. Each entry retains useCase, capability, reliability, value, ease, compatibility, overall, and rationale. These 1–5 editorial fit judgments are not quality reviews or measured benchmarks. An empty array means no Fit Scores are recorded. See the Fit Score methodology.externalRating: a third-party reference with publisher, rating, reviewCount, sourceUrl, and checkedAt, or null when no verified reference is available. It is not a TWF Review Score or an on-site review total.rating and reviewCount are retained unchanged where exposed, not silently redefined. legacyRating repeats their value and reviewCount, reports provenance (unverified or attributed-external-reference), and includes deprecated: true plus a notice. They are not TWF Review Scores or verified on-site review totals; use the separate review contract instead. No removal date is announced.score and matchedOn describe query relevance only. Relevance is neither a product-quality Review Score nor a use-case Fit Score. Relevance remains the primary ranking input; ties and empty queries use eligible product-quality reviews, then curated featured status and alphabetical name/slug. Unreviewed and limited entries are unscored, not zero-rated; legacy stars and review counts never rank tools.| review.status | Meaning |
|---|---|
not-reviewed | No published qualifying review; qualityScore and limitedScore are null. |
limited | A narrower published review; limitedScore may be present but qualityScore is null. |
reviewed | Current approved product-quality evidence supports qualityScore. |
needs-update | Material evidence needs refreshing; no current qualityScore is published. |
blocked | Review publication is held pending evidence; no qualityScore is published. Read reason. |
GET https://www.theworkflowfinder.com/api/v1/search
One endpoint, both content types. Pass type to search tools, workflows, or both at once (the default).
All parameters are optional. An empty query searches the full catalog, subject to the result limit. Tools use eligible product-quality reviews, then curated featured status and alphabetical order; legacy ratings and review volume are not ranking inputs.
| Parameter | Type | Description |
|---|---|---|
task | string (max 500) | Free-text description of what you're trying to do. Tokenized and matched against tool and workflow descriptions. |
category | string (max 50) | One of the site's 13 tool categories, e.g. Coding, Marketing, Design. |
pricing | enum | free, freemium, or paid. |
budget | number | Maximum monthly price. Excludes tools with a parseable price above it; tools with unparseable pricing (e.g. quote-based) are never assumed over or under budget, so they pass through. |
needs | comma-separated string | Extra free-text keywords to match, combined with task. |
avoid | comma-separated string | Keywords that hard-exclude a result if found anywhere in its text. |
capabilities | comma-separated string | One or more capability tags (see below). A tool matches if it carries at least one of the tags requested. |
openSource | enum | true or false. Filters on the tool's recorded open-source/self-hostable status. |
apiAvailability | enum | public, waitlist, none, or not-published. |
type | enum | tools, workflows, or both (default). |
limit | integer (1-50) | Max results per content type. Defaults to 10. |
capabilities, openSource, and apiAvailability filter against the same structured capability tags and source-linked facts published on individual tool pages: a privacy-policy URL, a vendor trust or security page where one is published, an open-source flag, API availability, and the date pricing was last verified. These are facts with a source, never scores.
Valid values for capabilities, grouped by branch:
Send commas between multiple values and URL-encode spaces, e.g. capabilities=Code%20generation,Code%20review. Not every tool carries capability tags yet; tagging is rolling out in traffic-ranked batches and currently covers most of the catalog.
Filter by capability and open-source status:
curl "https://www.theworkflowfinder.com/api/v1/search?capabilities=Code+generation&openSource=true&limit=1"{
"query": {
"capabilities": [
"Code generation"
],
"openSource": true,
"limit": 1
},
"tools": [
{
"slug": "aider",
"name": "Aider",
"url": "https://www.theworkflowfinder.com/tools/aider",
"category": "Coding",
"pricing": "free",
"pricingDetail": "Free (open source, bring-your-own API key)",
"shortDescription": "A terminal-based, Git-integrated AI pair-programmer that edits code across your whole repository using any LLM you configure. Free and open-source. You only pay your model provider directly.",
"rating": 4.6,
"reviewCount": 4100,
"review": {
"status": "not-reviewed",
"qualityScore": null,
"limitedScore": null,
"scale": 5,
"scope": null,
"testedAt": null,
"publishedAt": null,
"reviewUrl": null,
"reason": "No published, evidence-backed product-quality review yet.",
"methodologyUrl": "/review-methodology"
},
"fitScores": [
{
"useCase": "A terminal-first developer wanting AI coding help against their own local git repo without a new IDE or cloud upload",
"capability": 5,
"reliability": 4,
"value": 5,
"ease": 3,
"compatibility": 4,
"overall": 4.2,
"rationale": "Directly the tool's bestFor and standout: deep git integration that commits changes cleanly while working entirely with the local repository, matching pros on excellent git integration and being open source and self-hostable."
},
{
"useCase": "A team that already works in VS Code or Cursor and wants an integrated GUI experience",
"capability": 2,
"reliability": 3,
"value": 3,
"ease": 2,
"compatibility": 2,
"overall": 2.4,
"rationale": "The tool's own ourTake and con are direct: the command-line-only workflow is a real filter, and if a team works in VS Code or Cursor this won't fit naturally."
}
],
"externalRating": null,
"legacyRating": {
"value": 4.6,
"reviewCount": 4100,
"provenance": "unverified",
"deprecated": true,
"notice": "Legacy rating/reviewCount fields are not TWF Review Scores or verified on-site review totals."
},
"bestFor": "Developers who want AI coding help in the terminal against their own git repo, no new IDE or cloud upload required",
"ourTake": "Aider is for the terminal-first developer who wants full control and no new IDE. It runs entirely in your terminal, works against your local git repository, and commits changes cleanly as it edits. You bring your own API key from whichever model provider you prefer, so you're paying your provider directly rather than a subscription. That makes it genuinely free to run, with costs tied entirely to your actual usage. The command-line-only workflow is a real filter: if your team works in VS Code or Cursor, this won't fit naturally. For developers who live in the terminal and want git-integrated assistance without cloud lock-in, it's the right tool.",
"capabilities": [
"Code generation",
"Debugging assistance"
],
"facts": {
"privacyPolicyUrl": "https://aider.chat/docs/legal/privacy.html",
"openSource": true,
"apiAvailability": "none",
"pricingVerifiedAt": "2026-08"
},
"score": 3,
"matchedOn": [
"capability",
"openSource"
]
}
],
"workflows": [],
"meta": {
"totalToolMatches": 17,
"totalWorkflowMatches": 0,
"source": "https://www.theworkflowfinder.com",
"generated": "2026-08-28",
"scoreContractVersion": "1.0",
"scoring": {
"methodologyUrl": "https://www.theworkflowfinder.com/review-methodology",
"review": {
"statuses": [
"not-reviewed",
"limited",
"reviewed",
"needs-update",
"blocked"
],
"scale": 5,
"qualityScore": "Product-quality Review Score only when status is reviewed; otherwise null, never zero.",
"limitedScore": "A narrower published editorial rating, not a product-quality Review Score or a quality-ranking input.",
"nullMeaning": "No publishable product-quality score; not a negative quality judgment. See status, reason and scope."
},
"fitScores": {
"methodologyUrl": "https://www.theworkflowfinder.com/fit-score",
"scale": 5,
"meaning": "Existing use-case-specific editorial Fit Scores, not product-quality Review Scores or measured benchmarks. An empty array means no Fit Scores are recorded."
},
"externalRating": "Attributed third-party reference with publisher, sourceUrl and checkedAt; null when no verified reference is available. Not a TWF Review Score.",
"legacyRating": {
"deprecatedFields": [
"rating",
"reviewCount"
],
"notice": "Legacy numeric fields are preserved unchanged where exposed. Use legacyRating for their provenance and deprecation notice; they are not TWF Review Scores or verified on-site review totals."
},
"searchScore": "score is query relevance, not a Review Score or a use-case Fit Score.",
"ranking": "Query relevance first; ties and empty queries use eligible product-quality reviews, then curated featured status and alphabetical name/slug. Unreviewed and limited tools have no numeric quality rank; legacy ratings and review counts are never ranking inputs."
}
}
}Search workflows only:
curl "https://www.theworkflowfinder.com/api/v1/search?type=workflows&task=youtube+scripts&limit=1"{
"query": {
"task": "youtube scripts",
"type": "workflows",
"limit": 1
},
"tools": [],
"workflows": [
{
"slug": "produce-short-form-video-content-at-scale",
"title": "Produce Short-Form Video Content at Scale",
"url": "https://www.theworkflowfinder.com/workflows/produce-short-form-video-content-at-scale",
"difficulty": "Intermediate",
"timeEstimate": "2 hrs per week",
"description": "Repurpose one long-form video or podcast into 10+ short clips with captions, hooks, and platform-specific formatting, in under 2 hours per week.",
"expectedResult": "10–15 platform-ready short clips (Reels, TikTok, Shorts) with burned-in captions, strong hooks, and thumbnails, scheduled and ready to post for the week.",
"tags": [
"Video",
"Social Media",
"Content",
"Repurposing",
"Marketing"
],
"score": 2,
"matchedOn": [
"2 keyword matches"
]
}
],
"meta": {
"totalToolMatches": 0,
"totalWorkflowMatches": 28,
"source": "https://www.theworkflowfinder.com",
"generated": "2026-08-28",
"scoreContractVersion": "1.0",
"scoring": {
"methodologyUrl": "https://www.theworkflowfinder.com/review-methodology",
"review": {
"statuses": [
"not-reviewed",
"limited",
"reviewed",
"needs-update",
"blocked"
],
"scale": 5,
"qualityScore": "Product-quality Review Score only when status is reviewed; otherwise null, never zero.",
"limitedScore": "A narrower published editorial rating, not a product-quality Review Score or a quality-ranking input.",
"nullMeaning": "No publishable product-quality score; not a negative quality judgment. See status, reason and scope."
},
"fitScores": {
"methodologyUrl": "https://www.theworkflowfinder.com/fit-score",
"scale": 5,
"meaning": "Existing use-case-specific editorial Fit Scores, not product-quality Review Scores or measured benchmarks. An empty array means no Fit Scores are recorded."
},
"externalRating": "Attributed third-party reference with publisher, sourceUrl and checkedAt; null when no verified reference is available. Not a TWF Review Score.",
"legacyRating": {
"deprecatedFields": [
"rating",
"reviewCount"
],
"notice": "Legacy numeric fields are preserved unchanged where exposed. Use legacyRating for their provenance and deprecation notice; they are not TWF Review Scores or verified on-site review totals."
},
"searchScore": "score is query relevance, not a Review Score or a use-case Fit Score.",
"ranking": "Query relevance first; ties and empty queries use eligible product-quality reviews, then curated featured status and alphabetical name/slug. Unreviewed and limited tools have no numeric quality rank; legacy ratings and review counts are never ranking inputs."
}
}
}A malformed or out-of-range parameter returns 400 with the specific issue, not a generic failure:
curl "https://www.theworkflowfinder.com/api/v1/search?pricing=expensive"{
"error": "Invalid query parameters",
"details": [
{
"received": "expensive",
"code": "invalid_enum_value",
"options": ["free", "freemium", "paid"],
"path": ["pricing"],
"message": "Invalid enum value. Expected 'free' | 'freemium' | 'paid', received 'expensive'"
}
]
}Requests are limited per IP address to keep the endpoint available to everyone. Exceeding the limit returns 429 with a Retry-After header telling you how long to wait before retrying. There is no API key and no way to request a higher limit today; if you need sustained high-volume access, reach out via the contact page first.
The search API is one of several ways to consume this catalog programmatically:
POST /api/mcp: seven tools, no key, same response shapes as this API./workflows/[slug]/export.json: ordered steps, tool slots, and capability tags for that one workflow as a portable config.application/linkset+json) that lists this API, the MCP server, their descriptions and their documentation, and /.well-known/mcp/server-card.json is the MCP server card. Every page also sends Link headers pointing at these surfaces.This is a small site run by one person, not a funded API product. The endpoint has no formal uptime commitment, but it reads the same in-memory data every page on the site renders from, so it is only as unavailable as the site itself. Fields are additive over time as the catalog grows (new capability tags, new facts); existing fields won't be renamed or removed without notice here. If you build something on top of this, or find a bug in the response shape, let us know.