Workflow Automation With an API
As of August 2026, AIDiveForge tracks 35 workflow automation with an api. The top three by verified-data score are Coasty, MobileRun, and ego-lite. Curated workflow automation with an api tracked by AIDiveForge. Listings are verified against each tool's live website and re-checked regularly.
Last updated July 29, 2026 · 35 tools
Ranked by AIDiveForge's verified-data score: data completeness, verification recency, community rating, and real visitor engagement. How we rank · No tool can pay for placement.
1. Coasty
Coasty operates real software by sight: it browses portals, clicks, types, and verifies its own output without requiring an API from the target system. The vendor states it ranks first on OSWorld at 85.60% task completion, and the agent is designed to recover autonomously when a UI shifts — which is the specific failure mode that grinds traditional automation to a halt. It fits teams running long, multi-step workflows across carrier portals, ERPs, TMS platforms, or Microsoft 365 consoles. The ceiling appears when you need deep bidirectional data integration rather than screen-level execution; at that point, a native API connector will outperform it on speed and reliability.
Paid$99 flat Unlimited plan; $0.05 per /v1/predict API callAPIVerified Jul 10, 2026
2. MobileRun
The platform pairs an LLM agent — Claude, Hermes, or their own Mobilerun VA — with a persistent real device that has a real SIM, a real residential IP, and a real number receiving SMS. The agent reads screen state via accessibility tree plus vision, then taps, swipes, and types end-to-end. That identity stack is why apps like Instagram, WhatsApp, and Stripe behave the way they do in production rather than flagging the session. Fleet tooling handles parallel runs across up to 10,000 devices. The ceiling appears when workflows require branch logic the natural-language task description cannot express — at that point you are writing SDK code.
PaidOpen Source$5/monthAPISelf-hostedVerified Jul 9, 2026
3. ego-lite
ego (lite) is a custom Chromium build that installs as your daily browser and exposes a skill called ego-browser, which any code-writing agent — Claude Code, Codex, Cursor, Kiro — can drive directly. Agents run inside isolated Spaces so they don't collide with your open tabs. The vendor states the engine handles cross-origin iframes, shadow DOM, and third-party widgets like Stripe and Salesforce that JS shims typically fail on. The page claims task completion up to 3.45x faster than agent-browser tooling, on fewer tokens, because multiple in-page actions batch into a few lines of JavaScript instead of one tool call at a time. Mac-only at this point — Windows support is on a waitlist.
FreeOpen SourceAPISelf-hostedVerified Jul 28, 2026
4. Wayflow
The vendor describes Wayflow as an embeddable visual workflow editor — canvas, node palette, config panel, run controls, and a runtime engine that executes the same graph in the browser during development and on your server in production. A single call, createWorkflowEditor(), mounts the full workspace. Built-in node types cover LLM calls, tool-calling, branching, map-over-list, and image generation, so AI and deterministic steps sit on the same canvas. Suspend-and-resume is native, meaning a workflow can pause for a human review and pick back up without custom state management. The project is MIT licensed, carries zero runtime dependencies according to the docs, and ships full TypeScript types.
FreeOpen SourceAPISelf-hostedVerified Jul 8, 2026
5. Browser-Use
The open-source library installs via pip and gives an LLM-controlled browser the ability to complete multi-step tasks — form filling, navigation, data extraction — with a self-healing loop that recovers from DOM changes without manual patching. The cloud tier adds anti-detect browsers, CAPTCHA solving, and residential proxies across 195+ countries, all zero-config. The vendor reports 97% benchmark accuracy on browser tasks, and cold start times are described as sub-second. At production scale, the hosted agent layer handles parallel sessions, so the constraint shifts from infrastructure to prompt quality and LLM cost. Teams running high-frequency extraction at scale will hit the point where LLM call costs per task matter more than the browser infrastructure fee.
PaidOpen Source$29/moAPISelf-hostedVerified Jun 30, 2026
6. BrowserAct
The core loop is prompt-in, structured-data-out: describe what you need, the agent builds and tests a Bot, then publishes it as a reusable scraper you can trigger from Make, n8n, or Zapier. Built-in residential proxies and CAPTCHA handling mean protected pages are reachable without assembling your own infrastructure. The local agent option lets teams run browsers on their own hardware when data cannot leave the building. The ceiling appears when your extraction logic requires conditional branching across multiple page types — the prompt interface has no canvas for that, so complex workflows still need a surrounding orchestration layer. Community ratings on G2 sit at 4.6, suggesting the core promise holds for straightforward collection tasks.
PaidOpen SourceFree Trial · 7 daysAPISelf-hostedVerified Jul 29, 2026
7. OpenIngress
OpenIngress is an open-source, self-hosted tool that scans public-facing websites and reports where AI agents will break: unlabeled interactive elements, flow blockers before key actions, and UI patterns that confuse automated navigation. The core workflow is pointed at a URL — the tool crawls the target and surfaces specific compatibility issues along with fix recommendations aimed at developers. It fits cleanly into pre-deployment audits for e-commerce flows, form submissions, or any checkout-critical path. The self-hosted model means scan results stay on your infrastructure. The scrape-and-report approach covers what's publicly visible, which is the ceiling — authenticated flows and dynamic states behind login require additional tooling.
FreeOpen SourceAPISelf-hostedVerified Jul 23, 2026
8. Reflekt Ai
Reflekt reads incoming emails and their attachments, extracts order or invoice line items using OCR and AI, and hands a fully prepared ERP record to a human for one-click approval before it posts. The vendor states that SMEs processing several thousand documents per year hit a clear ROI case. The approval gate stays in place by design — this is not a system that posts records autonomously, which matters for teams where a bad line item means a wrong shipment. ERP connection is API-based, so the tool works alongside your existing system rather than replacing it. The ceiling appears when document volumes or ERP complexity grow beyond what a single-inbox, approval-first model can absorb without bottlenecks.
PaidAPIVerified Jul 27, 2026
9. Teable 3.0
Teable positions itself as an AI-native database that lets you describe what you need in plain language and get structured tables, automations, and basic apps without writing code. File ingestion works well for structured extraction tasks — receipts, contracts, resumes — where the AI fields parse and populate rows automatically. The automation layer handles triggers and actions for teams that have outgrown Zapier-style one-step rules but are not ready to maintain a full workflow engine. The ceiling appears when logic gets complex: branching conditions and multi-table orchestration push past what the chat interface can express cleanly. Teams hitting that wall typically bolt on a separate scripting layer or migrate to a purpose-built backend.
PaidAPISelf-hostedVerified Jul 24, 2026
10. Tsaagan
The architecture centers on perception-action-verification loops rather than fire-and-forget scripting, which means each browser action waits for confirmed state before the agent proceeds. Tsaagan ships an MCP server alongside JS and Python SDKs, so agents already wired into those runtimes can call browser actions without building a separate automation layer. It runs on Playwright, native APIs, and a browser extension — giving it reach across sites that block headless fingerprints. The public repo shows 29 commits and three open issues, which signals early-stage software; production teams should expect rough edges and plan to contribute fixes. For simple, authenticated scraping pipelines it earns its place — for high-volume, concurrent agent fleets the maturity ceiling appears quickly.
FreeOpen SourceAPISelf-hostedVerified Jul 22, 2026
11. AgentX
The platform covers three stages most tools split across separate vendors: a drag-and-drop canvas for multi-agent workflows, a built-in eval layer where you run agents against test sets before shipping, and one-click deployment to API, Slack, web widget, email, or voice. Versioned deploys with rollback and per-run logs mean incidents are traceable rather than mysterious. For agencies, white-label plans let you ship agents under your own brand. The visual builder will hit its ceiling when branching logic grows complex enough to require conditional depth the canvas cannot cleanly express — at that point teams are either rearchitecting or adding code outside the builder. There is no self-hosted option, so teams with data-residency requirements are blocked before they start.
Paid$49/moAPIVerified Jun 28, 2026
12. Atlas
The platform layers document extraction, a rule-encoding structure the vendor calls a context graph, exception handling, and ERP posting into a single agent loop — so invoices that arrive in any format get validated against POs, routed for approval, and posted without a person in the middle. The vendor states their OCR model ranks first on the IDP Leaderboard, ahead of GPT-5, Gemini, and Claude. Where the system earns its keep is exception resolution: when a field doesn't match, the agent checks it against your encoded rules rather than dropping it in a queue. Every decision traces back to the rule and document that drove it, which matters when an auditor asks.
PaidAPISelf-hostedVerified Jun 27, 2026
13. ComfyUI
ComfyUI exposes every model, sampler, conditioning step, and post-processing node as a discrete block on an infinite canvas, so the pipeline that produced the result is the documentation. VFX teams and creative studios use it to build multi-model pipelines where every parameter is auditable and every step can be swapped independently. The node graph runs locally via Comfy Desktop or in the cloud, and workflows can be published as API endpoints through Comfy API. Where it breaks: the node graph is not a gentle on-ramp. Teams that need non-technical colleagues to run pipelines use App Mode, a simplified overlay, but deep customization still requires fluency with the underlying graph.
PaidOpen SourceAPISelf-hostedVerified Jun 30, 2026
14. LiveContext
LiveContext combines chat-driven workflow building, multi-step agents with credit budgets, an auditable execution log, and a marketplace where teams can share or fork apps. The self-hosted option means organizations with data residency requirements are not forced onto a vendor cloud. Agents run tasks on their own — support email triage, expense approvals, lead normalization — and the credit budget system gives you a hard ceiling on runaway execution costs. The scraped page content is thin on specifics, so precise limits on step counts, integration depth, and throughput are not sourced; the architectural pattern is confirmed but the edge cases are not documented publicly.
PaidAPISelf-hostedVerified Jul 7, 2026
15. Webskillet
Webskillet runs browser tasks through an agent that figures out the extraction logic on run one, then saves that logic as a reusable 'Skillet' — a code path that replays cheaply and self-heals if the target site drifts. The vendor states this makes costs drop sharply after the initial run, which matters when you're hitting the same endpoint daily. You define an output schema upfront, the agent handles the browsing, and the result comes back structured. The ceiling appears when a task requires judgment calls that the first run didn't anticipate — novel page states, authentication flows, or sites that actively block headless browsers. Cloud-only execution means you cannot route traffic through your own infrastructure.
PaidAPIVerified Jul 23, 2026
16. Airparser
Airparser takes unstructured documents — emails, PDFs, scanned forms, handwritten notes — and pulls structured fields out of them using GPT-based extraction rules the user defines. The workflow is: import a document, describe what fields you want, and the engine returns a clean JSON or CSV you can route into Google Sheets, a CRM, or a downstream automation. It holds up well for finance teams processing consistent invoice formats and HR teams ingesting CVs at volume. The ceiling appears when document layouts vary enough that a single extraction schema stops covering all variants — teams end up maintaining multiple schemas rather than one. Documents that require cross-referencing data across pages or multi-table reconciliation push outside what the extraction model reliably handles.
PaidFree Trial · 30 days$33/moAPIVerified Jun 3, 2026
17. CartAI
CartAI exposes a single API endpoint that dispatches an agent to navigate any web property and complete a transaction: checkout, subscription sign-up, invoice payment, vendor portal submission. The workflow is async — you POST a task with customer, payment, and shipping profiles, then listen for webhook events (STARTED, COMPLETED, FAILED) and retry from the failure signal. The vendor describes composable step primitives with conditional branching and idempotent, retry-safe execution, which means a failed 3DS step doesn't silently drop the order. The constraint is intentional: CartAI is built exclusively for workflows where a transaction has to clear, not for scraping, research, or general navigation. That focus is the architectural bet — teams needing general browser automation will hit the ceiling immediately.
PaidAPIVerified Jun 26, 2026
18. Creativly.ai
The workspace covers image generation, video production, audio, text, and a visual workflow builder under one login, pulling from models like FLUX, Kling, Veo, Sora, Runway, and GPT Image via Replicate, WaveSpeed, and Gemini. The Flow builder lets you wire multi-step creative pipelines — generate a hero shot, spin color variants, assemble a storyboard grid, export a product video — without leaving the platform. Pre-built templates for UGC skincare campaigns, sneaker drop ads, and virtual try-on workflows mean agencies skip the blank-canvas setup. The credit model works alongside bring-your-own-API-key, so teams with existing OpenAI or Replicate accounts avoid double-paying. The ceiling appears when a workflow needs logic that branches based on conditional output — the visual canvas handles linear chains, not branching trees.
Paid£20/month or £30/monthAPIVerified Jun 9, 2026
19. cua
Cua provisions cross-OS fleets from a single API, forks machine state over copy-on-write snapshots so you can reproduce failures without rebuilding from scratch, and serves pre-booted machines from warm pools that claim in milliseconds. The open-source Cua Driver runs background desktop automation on macOS and Windows — agents click, type, scroll, and inspect accessibility trees without stealing your cursor. Linux support in Cua Driver is in pre-release, so teams with Linux-heavy desktop workflows will hit that wall immediately. At scale, you either point your training loop at live warm pools or order verified trajectory datasets that arrive pre-packaged for your ingestion pipeline.
PaidOpen SourceAPISelf-hostedVerified Jun 18, 2026
20. DataDack
The platform runs visual workflow orchestration, AI agents with RAG memory, and IoT telemetry ingestion under one roof, deployed on AWS Mumbai and Hyderabad for teams that cannot let data cross Indian borders under DPDP. The vendor states 10ms node latency and a 99.9% uptime target at 10k+ RPS — claims that hold architectural credibility given the Go and Node.js core, but production verification at your specific load profile is still your job. The agent builder and RAG memory features are paid-only. Teams on the free tier get workflow automation and gateway access, but the autonomous swarms stay behind a paywall.
Paid₹1,799/mo INR or $19/mo USD (Starter tier minimum paid)APIVerified Jun 18, 2026
21. Docunerve
Docunerve accepts PDFs — including scanned documents — and returns structured Markdown or JSON that downstream LLM pipelines can actually consume. The vendor states it handles multilingual documents and preserves tables, formulas, and layout structure that generic parsing libraries flatten or drop. For teams running high-volume ingestion into vector databases, the API-first design means extraction slots into existing pipelines without a UI bottleneck. The ceiling appears when your documents demand post-extraction logic, conditional routing, or validation steps — Docunerve performs one-shot extraction and stops there. Teams with more complex orchestration needs wire the output into a separate processing layer.
Paid$0.01/pageAPIVerified Jun 12, 2026
22. DoMyWork
The tool operates in two modes: Chat, where you issue a task and the agent executes it end-to-end, and Autopilot, where recurring tasks run on a schedule without you touching anything. Lead enrichment, competitor price tracking, and report generation are the documented sweet spots — tasks where the inputs are structured and the output format is predictable. The agent executes code and API calls autonomously, which means it handles multi-step sequences without a node-by-node canvas. The ceiling appears when tasks require complex conditional branching or when output quality depends on edge cases the agent hasn't been prompted to handle — at that point, teams fall back to manual prompt tuning or external scripting.
Paid$15.99/moAPIVerified Jun 8, 2026
23. Freu AI
Freu AI's approach is observe-once, compile, execute-forever: a human performs a workflow, the agent records and compiles it into a locally-runnable program, and from that point forward execution runs without calling a model on every step. The vendor positions this as the core cost argument — token spend happens during the learning phase, not during the thousands of subsequent runs. That architecture fits invoice routing through ERPs, clinical evidence extraction, and batch record migration across legacy systems that have no API surface. The wall appears when a workflow changes: any meaningful UI or process shift requires a new learning pass, which means ongoing human expert time isn't eliminated, just front-loaded.
PaidToken-based learning cost + free executionAPISelf-hostedVerified Jun 1, 2026
24. GhostUser
Each persona — a cautious newcomer, a skeptical evaluator, a power user, a time-pressured visitor, a motivated buyer — navigates your app autonomously, flags where it gave up, and logs why. Console errors, failed network requests, and 5xx responses get caught in the same pass, so you get UX feedback and QA signal in one run. It connects directly to localhost, which means you catch issues before they leave your machine. The tool runs on your Claude API key, so cost scales with usage rather than with a seat count. Where it breaks: the feedback reflects what five hardcoded personas notice, not the distribution of your actual users.
FreeOpen SourceFree (pay-per-token for Claude API calls)APISelf-hostedVerified Jun 9, 2026
25. Gumloop
Gumloop lets growth, sales, and ops teams wire together multi-step AI agents that run on their own — pulling from external APIs, enriching CRM records, drafting content, and firing results into Slack or Teams without a human trigger per run. The visual builder handles the common cases well: lead enrichment, meeting prep, competitive research. Branching logic that depends on what a previous step returned is where the ceiling appears — complex conditional paths push teams toward adding custom code nodes, which means they are now maintaining two layers. Security and compliance teams get enterprise-grade controls over AI usage, which matters when rolling out to non-technical employees at scale.
Paid$37/monthAPISelf-hostedVerified Jun 4, 2026
26. HARPA AI
The extension activates on any webpage via a keyboard shortcut and surfaces contextual AI actions tied to what's on screen — summarize this thread, draft a reply in your tone, extract this table, monitor this price. Web automation tasks like form-filling, data scraping, and page-change alerts run without you staying at the keyboard. The privacy architecture is the real differentiator: conversations are not logged by the vendor, local models are supported, and GDPR compliance is vendor-stated. The ceiling appears when automation sequences grow complex — multi-step conditional flows that depend on dynamic page states push against what the extension model can reliably handle. Teams building more than simple linear automations typically reach for a dedicated orchestration layer alongside it.
PaidS2 Plan costs $19 per monthAPIVerified Jun 9, 2026
27. Infer0
infer0 sits between your app and the AI provider: users connect their OpenAI, Anthropic, or Google keys, authorize your app via OAuth, and infer0 forwards requests while translating between API formats so your existing SDK calls work unchanged. Your app never touches a key. Spend limits live on the user side, enforced per-provider and per-authorization, revocable in one click. The architecture is passive middleware — no agent logic, no workflow builder — which means it integrates cleanly but covers only the routing and auth layer. If infer0 goes down, your app's requests fail; the docs are explicit: handle that gracefully.
FreeAPIVerified Jun 18, 2026
28. Make (Integromat)
Make lets you build automation sequences by dragging operations onto a canvas—no coding required. You're essentially replacing repetitive work (data entry, email sends, syncing spreadsheets to CRMs) with conditional logic that runs on schedule or trigger. The free tier includes 1,000 operations per month; paid plans start around $10/month for 10,000 operations. The honest catch: the free plan's operation limit exhausts quickly for serious workflows, and the visual builder can feel clunky for genuinely complex logic compared to traditional automation code.
Paid$9.99/moAPIVerified Dec 1, 2025
29. Nextqore
Because the factual source and the tool metadata describe entirely different products, generating accurate production-reality content for this listing is not possible without verified, on-topic source material. Publishing listing content drawn from the wrong vendor page risks misinforming engineering leads and product managers who are making real infrastructure decisions. The structured data describes a paid SaaS data preprocessing and lineage platform targeting teams running agentic AI systems at scale — a product that deserves accurate, grounded copy. No claims about Nextqore's Spotter can be sourced from the provided page, and fabricating capabilities would violate the grounding rules of this system. This listing should be held until the correct vendor page is supplied.
Paid$1,200/moAPIVerified Jun 4, 2026
30. Onpilot
The platform connects agents to ERP, CRM, support tools, and custom APIs, then layers in approval steps, permission scopes, and audit logs so the agent cannot act unilaterally on sensitive operations. Agents can search, reason, take action, and hand off to a human — the approval step pauses execution and sends an interactive Slack message before anything ships. Multi-tenant architecture means a single deployment can serve isolated customer or plant workspaces with per-tenant access control. Where it breaks: Onpilot is a custom-built, consultative engagement, not a self-serve platform you configure over a weekend — teams without clear workflow documentation will stall during scoping.
PaidAPISelf-hostedVerified Jun 12, 2026
31. Ornold MCP
The structured data describes a browser automation platform for parallel antidetect workflows, vision-first interaction, and CAPTCHA solving at scale. However, the scraped page content is from an unrelated travel-identification app called Spotter. There is no factual basis from the page to describe how the tool handles parallel execution, how its AI agent layer interprets natural-language task definitions, where its CAPTCHA solving hits rate limits, or when the free tier stops being sufficient. Publishing claims without a sourced page would mean fabricating production details — the one thing an engineering lead or PM cannot afford to act on.
Paid$0/moAPISelf-hostedVerified Jun 2, 2026
32. Relay
Relay.app lets you describe a workflow in plain language, then generates a visual step sequence you can edit manually or by prompting again. The core model is fixed-sequence automation — triggers, steps, branches, loops — with AI inserted at specific points for extraction, summarization, or creation, not for deciding what to do next. Approval gates are built in, not bolted on, so a finance director can sign off on an expense before it routes to payment. Reusable 'Sequences' let teams standardize common patterns like lead enrichment or onboarding and propagate updates across every workflow at once. The ceiling appears when logic grows complex: deep conditional branching across many steps pushes against what the visual canvas expresses cleanly.
Paid$19/moAPIVerified Jun 1, 2026
33. SoMatic
The core workflow is a CLI command that takes a screenshot, runs element detection locally, and returns numbered marks with coordinates as JSON — so agents target elements by ID, not by fragile pixel hunts. Every action returns JSON, which means downstream agents can chain steps without parsing unstructured output. The self-hosted, MIT-licensed model runs on your own hardware, so no screenshot data leaves the machine. The wall appears with non-standard or highly dynamic UIs where YOLO detection misses elements or mislabels them — teams handling those cases add a fallback coordinate layer manually. At this GitHub star count, the community size is small, which means debugging edge cases happens in the codebase, not a forum.
FreeOpen SourceAPISelf-hostedVerified Jun 9, 2026
34. Vessel Browser
Vessel exposes 80+ browser tools over MCP — navigate, click, fill, extract, diff, checkpoint, undo — so any MCP-compatible agent harness can drive a real browser without brittle Playwright scripts or headless guesswork. The visible-by-default design means you can watch, pause, approve, or roll back before something ships. Session persistence and named checkpoints mean the agent picks up where it left off instead of re-authenticating from scratch every run. The wall appears on complex, multi-site workflows that branch based on what the last page returned — the built-in skill system handles linear recipes well, but conditional logic across domains requires more than reusable instructions. Saved sessions, credential vaults, and page-change history are paid-only features, so teams running fully autonomous monitoring pipelines hit that ceiling fast.
PaidOpen SourceFree Trial · 7 days$5.99/monthAPISelf-hostedVerified Jun 19, 2026
35. Zapier AI
Zapier sits between your apps—Slack, Gmail, Salesforce, etc.—and lets you create if-this-then-that automations (called Zaps) that trigger actions across platforms. The core problem it solves is manual data entry and context-switching; instead of copying information between tools, Zapier does it. Pricing starts free for basic testing, then $19–$299/month for individuals and teams, scaling sharply for enterprises. The trade-off is clear: simple automations work beautifully, but complex multi-step logic and edge cases often require premium tiers or workarounds, making it expensive for organizations running hundreds of Zaps.
Paid$19.99/moAPIVerified Dec 1, 2025
Listings on this page are sourced and verified by the AIDiveForge data pipeline. AIDiveForge is editorially independent — no money changes hands for inclusion.