Keystroke: the address we list did not respond · last checked 8 Nov 2026
Keystroke
Build internal AI agents as deployable TypeScript — from chat, Cursor, Claude Code, or MCP
Keystroke is the pick when you want reasoning agents with sandboxed code and durable workflows, but insist the artifact be TypeScript you can review and own. The metered $0.01-per-run model, with compute and search passed through at cost, is honest and predictable. Weigh the caps before committing — timeouts and concurrency are limited on Hobby and Pro.
Verdict generatedLast checked 29d ago · cite: rightaichoice.com/tools/keystroke
- Engineering teams that want agents reviewable as TypeScript in their own repo
- Startups shipping internal automation on metered, usage-only pricing
- Teams already using Cursor or Claude Code that need a deploy target
- Ops, sales, and finance teams with a technical partner nearby
- Teams that need a purely visual, no-code automation builder
- Solo users who just want a simple chatbot
- High-volume, customer-facing workloads that exceed Pro's concurrency limits
We scan live Reddit threads, YouTube comments, X posts, G2 reviews and other communities — and hand you an honest verdict in under a minute.
- Honest verdict, not marketing
- Real pros & cons from real users
- Attributed quotes with receipts
3 free scans · no card needed
Skip Keystroke if you need a purely visual, no-code builder, run high-volume customer-facing workloads above 10 concurrent runs, or need SSO and audit logs without moving to the Organization plan.
Going past Pro's 30-minute agent or workflow timeout means the run is cut off, and longer jobs need the Organization plan's custom timeout.
Keystroke is cheap to start and cheap to experiment on: Hobby is free forever with a $1/mo usage credit and 1 project, and Pro is $20/mo plus additional usage with $20 of included credit, 5 projects, and 10 concurrent runs. That undercuts most managed agent platforms for small internal automation, and it beats writing your own plumbing when you factor in the included memory, sandbox, integrations, and traces. Costs scale with actual run volume rather than seats, which favors small teams doing a
In short
Keystroke — Build internal AI agents as deployable TypeScript — from chat, Cursor, Claude Code, or MCP. Best for Engineering teams that want agents reviewable as TypeScript in their own repo, Startups shipping internal automation on metered, usage-only pricing, Teams already using Cursor or Claude Code that need a deploy target. Free to start; paid plans from $20/mo.
What is Keystroke?
Keystroke — Build internal AI agents as deployable TypeScript — from chat, Cursor, Claude Code, or MCP. Best for Engineering teams that want agents reviewable as TypeScript in their own repo, Startups shipping internal automation on metered, usage-only pricing, Teams already using Cursor or Claude Code that need a deploy target. Free to start; paid plans from $20/mo.
Keystroke is an AI agent platform where the deliverable is plain TypeScript in your own repository, not a locked visual flow. You build in one of three ways: a platform agent writes, tests, and deploys the code from a chat prompt (no IDE needed), you point Cursor, Claude Code, or Codex at Keystroke's open framework and headless CLI, or you connect over MCP from ChatGPT or Claude. All three routes write to the same project, so engineers and non-engineers share one codebase. The agents themselves get real tooling: persistent memory across conversations, a workspace with filesystem, bash, and isolated VM execution, database access, web search via Exa, and a browser that stays signed in between runs. Durable workflows layer on step replay, typed Zod input and output, durable sleeps, and human approval hooks. Triggers fire from schedules, webhooks, or events inside connected apps. Reach is wide — 1,000+ integrations including Slack, Microsoft Teams, Telegram, Discord, Linear, and WhatsApp, plus any HTTP API, any MCP server, and 300+ models through an AI gateway with bring-your-own-key. Run traces show what happened across agents, tools, and human steps, and agents can spin up instant apps for dashboards and internal tools. Metering is on execution: $0.01 per agent run, $0.01 per workflow run, $0.005 per empty poll trigger, with sandbox compute and Exa calls passed through at cost. Teams that want automation surfacing in Slack or Teams rather than another chatbot tab — and that would rather review code than drag boxes — are the natural fit.
Key features
- Platform agent that writes, tests, and deploys agents as TypeScript from chat
- Open TypeScript framework and headless CLI for Cursor, Claude Code, and Codex
- MCP connection to build agents from ChatGPT, Claude, or any MCP client
- Agent memory that persists across conversations
- Persistent workspace with filesystem, bash, and isolated VM code execution
- Database access for agents
- Web search via Exa and web fetch
- Persistent browser that stays signed in across runs
- Durable workflows with step replay and typed Zod input and output
- Durable sleeps and human approval hooks
- Triggers from schedules, webhooks, and connected-app events
- 1,000+ integrations plus any HTTP API and any MCP server
- External channels including Slack, Microsoft Teams, Telegram, Discord, Linear, WhatsApp
- Instant apps for agent-built dashboards, websites, and internal tools
- Run traces across agents, tools, and human decisions
Behind the verdict
Most agent builders ask you to trust a canvas. Keystroke asks you to read a pull request. That single design choice is why we'd shortlist it for engineering teams and quietly skip it for everyone else. Pick Keystroke when the automation needs to live where work already happens — a Slack channel, a Teams thread, a Linear ticket — and when someone on the team will actually open the repo. The TypeScript output means code review, versioning, and rollback all work the way your team already works. The three entry points (chat, Cursor/Claude Code via the open framework and headless CLI, or MCP from ChatGPT) matter more than they sound: a non-engineer can prototype in chat and an engineer can take the same project over without a migration. Pass if nobody on the team wants to touch code. A purely visual builder will be faster to a first automation, and Keystroke's value — reviewable source, durable replay, approval hooks — is largely wasted on a team that never reads it. Solo users who want a chatbot are also the wrong audience; this is infrastructure, priced like infrastructure. Where it bites: Hobby and Pro carry timeouts and concurrency caps that will surface on long batch jobs and high-volume customer-facing traffic. The metered model is fair, but it is still metered — sandbox compute and Exa calls pass through at cost, so a chatty agent doing lots of search is not a flat bill. Budget a month of real usage before you promise anyone a number. The Organization plan lists a Company Brain — a shared space for teams to collaborate on context, agents, and skills — as coming soon, with no release date published. Treat it as roadmap, not a purchasing reason. Compare against visual automation suites when the buyer is ops-only, and against raw framework work when you have
Models under the hood
as of 2026-10-15
Limitations
- Everything you build is real TypeScript in your own project, so you need comfort with repos, git, and tests — or you lean on the platform agent or MCP to write, test, and deploy it.
- Plan limits are the practical ceiling: Hobby allows 1 project, 5 concurrent runs, 5 GB storage, 60 requests/min on API and MCP, 7-day log retention, and a 10-minute agent and workflow timeout; Pro raises these to 5 projects, 10 concurrent runs, 50 GB storage (then $0.25 per 10 GB), 300 requests/min, 30-day logs, and 30-minute timeouts.
- Governance features — role-based access controls, agent guardrails, advanced usage and spend limits, unlimited concurrency, SSO, audit logs, and enterprise-grade security — are Organization-plan only.
- Usage is metered while work runs: $0.01/agent run, $0.01/workflow run, $0.005/empty poll, Exa search $0.007 and fetch $0.001 at cost, sandbox compute about $0.067/hour at cost, and a 1.1x markup on gateway models if you don't bring your own keys.
- Company Brain is listed as coming soon with no published release date.
as of 2026-10-09
Researching Keystroke? Get your full AI stack in 60 seconds.
Free, no signup — tell us your goal and get tools matched to your budget & existing stack.
Is Keystroke still active?
Keystroke: the address we list did not respond · last checked 8 Nov 2026.
Viability score
How well maintained and how widely used is Keystroke? Built from what the vendor actually publishes (docs, changelog, tutorials, integrations, pricing), whether the site is live, and how much real users discuss it. How we calculate this
Last calculated: November 2026
How we score →What do users say about Keystroke?
What people actually say about Keystroke — is it worth it?
Our community sentiment research written Aug 6, 2026 is more than 90 days old, so it is being re-reviewed against current data and is not shown.
How we verified this
Every listed fact about Keystroke — pricing, features, links, status — is re-read from the vendor's own pages by our verification pipeline and each change is logged with a date; the last pass ran on 16 Oct 2026. Analysis is drafted with AI models and checked against our editorial standards before publishing; see how we verify.
Verification history
We have re-verified Keystroke 7 times since . Each pass re-reads the vendor's own pages and re-checks every listed field against that evidence; passes where nothing had changed are marked as such.
- — changed summary, pricing tiers, description, who should skip it, our verdict
- — changed link health, website status
- — changed SSL certificate
- — changed features, who should skip it, summary, description, pricing tiers, our verdict
- — changed SSL certificate
- — changed who should skip it, switching out, models, description, summary, our verdict, hidden costs, setup time, pricing fit, limitations, switching in
Showing the 6 most recent of 7 verification passes.
Free to cite with attribution — this page re-verifies continuously.
How much does Keystroke cost in 2026?
Keystroke: Free to start; paid plans from $20/mo. Pricing last verified 16 Oct 2026.
Plans compared
Every published Keystroke tier, its price as the vendor publishes it, who it fits, and what it adds over the tier below. The 12-month cost projection follows.
| Plan | Price | Billing | Who it's for |
|---|---|---|---|
| Hobby | Free | — | A solo builder or small team wanting to explore Keystroke with one project, 5 concurrent runs, 5 GB storage, 7-day logs, and a $1/mo usage credit before committing to a paid tier.Adds: Starting tier — free forever, deploys only from your own repo, 1 project, 5 concurrent runs, 60 requests/min on API and MCP, 10-minute agent and workflow timeout. |
| Pro | $20/moVendor: “$20/mo + additional usage” | monthly | A team running several internal agents in production that needs $20/mo of included usage credit, 5 projects, 10 concurrent runs, 50 GB storage, 30-day log retention, and 30-minute timeouts.Adds: Adds $20/mo of included usage credit, 5 projects instead of 1, 10 concurrent runs, 300 requests/min on API and MCP, 50 GB storage (then $0.25 per 10 GB), 30-day logs, and 30-minute timeouts. |
| Organization | Custom | quote | Companies that need role-based access, SSO, audit logs, advanced spend limits, agent guardrails, and unlimited concurrency for regulated or security-reviewed deployments.Adds: Adds advanced usage and spend limits, role-based access controls, agent guardrails and controls, unlimited concurrency, plus the security tier features; Company Brain is listed as coming soon. |
12-month cost
Project the real annual outlay, including the implied monthly cost when only an annual tier is published.
Vendor list price only. Add-on usage, seat overages, and contract minimums are surfaced under Hidden costs & gotchas.
Where the pricing makes sense
The company stage and team size where Keystroke's pricing actually pencils out — and where peers do it cheaper.
Keystroke is cheap to start and cheap to experiment on: Hobby is free forever with a $1/mo usage credit and 1 project, and Pro is $20/mo plus additional usage with $20 of included credit, 5 projects, and 10 concurrent runs. That undercuts most managed agent platforms for small internal automation, and it beats writing your own plumbing when you factor in the included memory, sandbox, integrations, and traces. Costs scale with actual run volume rather than seats, which favors small teams doing a
Has Keystroke's pricing changed?
Keystroke's pricing is tracked for changes; the moves we have logged since Aug 2026 are below.
| Date | Change |
|---|---|
| Pro: $29 → $20 /mo | |
| Plan renamed: Free → Hobby (price unchanged) | |
| Plan renamed: Enterprise → Organization (price unchanged) | |
| Plan removed: Team |
Who should pick Keystroke?
Pick Keystroke if you are Engineering teams that want agents reviewable as TypeScript in their own repo, Startups shipping internal automation on metered, usage-only pricing, Teams already using Cursor or Claude Code that need a deploy target, Ops, sales, and finance teams with a technical partner nearby.
Look elsewhere if you are Teams that need a purely visual, no-code automation builder, Solo users who just want a simple chatbot, High-volume, customer-facing workloads that exceed Pro's concurrency limits, Long batch jobs that run past the Pro agent or workflow timeout.
Skip it if Skip Keystroke if you need a purely visual, no-code builder, run high-volume customer-facing workloads above 10 concurrent runs, or need SSO and audit logs without moving to the Organization plan.
GeneratedReal-world workflow fit
Concrete scenarios for the personas Keystroke actually fits — and what changes day-one when you adopt it.
Open the Keystroke web app, ask the platform agent to build an agent named Timmy wired to Superhuman tools (list/search threads, get thread, draft, send, labels), set the guardrail to draft-by-default, then connect a Slack gateway so the team can request triage on the go.
Outcome: Inbox triage and reply drafting run as a deployed TypeScript agent in the team's project, with an inspection trace showing which tools fired and which drafts a human approved.
Point Claude Code at the Keystroke repo, build `src/agents/timmy.ts` with typed Zod input and output, then run `keystroke typecheck`, `keystroke test`, and `keystroke deploy --filter agents/timmy` from the terminal before prompting the deployed agent with `keystroke agents prompt timmy --wait`.
Outcome: The agent lands as reviewable TypeScript in the existing repo with an integration test, so the change goes through normal code review instead of living in a separate automation tool.
Build a support-desk agent, wire the tool actions it needs, and run the eval suite (the docs example shows 12/12 cases passing against a Support Agent) before flipping it on, then route status updates into Slack via an external channel.
Outcome: The agent ships with a documented pass rate instead of shipping on vibes, and support requests are handled where the team already works.
Use cases
- Triage an inbox and draft replies in your voice using an agent wired to Superhuman mail tools
- Turn a merged pull request into a shipped-work summary via a trigger on the main branch
- Confirm whether pricing was agreed on a customer call and post the answer to Slack and Linear
- Automate a support desk with an agent you run evals against before production
- Spin up an instant internal dashboard for pipeline, revenue, and team performance
- Build a morning briefing that pulls from connected apps on a schedule
- Give every team one place to share agents, skills, and company context
Setup time & first value
How long it actually takes to get something useful out of Keystroke — broken out by persona, not the marketing-page minute.
With the platform agent in the web app, you can have a first deployed agent running in a single chat session — the docs walk through building, testing, and deploying without touching an IDE. From an existing repo with Cursor or Claude Code, budget an afternoon: install the CLI, write one agent file, run `keystroke typecheck` and `keystroke test`, then `keystroke deploy`. Connecting OAuth
Switching to or from Keystroke
How to bring data in from common predecessors and how to get it back out — written for the switcher, not the buyer.
- →From n8n or Windmill: rebuild the flow as TypeScript agents and durable workflows, then move triggers from schedules and webhooks into Keystroke's trigger system.
- →From a hand-rolled agent stack: keep your model calls but replace your memory, sandbox, and credential plumbing with agents, workspaces, and the credentials store.
- →From ChatGPT or Claude chat automations: connect Keystroke over MCP and build from the same chat surface, but land the output as code in your project.
- →From Zapier-style task automation: map each task to an agent tool call or workflow step, then add run traces and evals that a task runner doesn't give you.
- ↗To a visual builder like n8n or Windmill: port agents into node graphs, accepting that step replay and typed Zod input/output have no direct equivalent.
- ↗To a self-hosted agent framework: export the TypeScript you already wrote and re-implement memory, sandbox, and credential handling yourself.
- ↗To a managed agent platform with seat-based pricing: rebuild triggers and channels, and re-price based on seats rather than metered runs.
- ↗To a homegrown deployment: keep the Keystroke TypeScript and replace the runtime, run history, and trace views with your own observability stack.
Integrations
Tutorials & Learning
YouTube returned 6 videos for “Keystroke”, and we withheld 6: 6 could not be judged, because “Keystroke” is a single word that other videos use for other things. We are showing none, because we could not prove any of them are about Keystroke.
Official links
What are the best Keystroke alternatives?
The closest alternatives to Keystroke we list are n8n, Openbot and AutoGen Studio. Ranked by direct product-type match, not category overlap.
| Alternative | Pricing | Starts at | Last verified |
|---|---|---|---|
| n8n | Freemium | Pricing not published | — |
| Openbot | Freemium | Pricing not published | — |
| AutoGen Studio | Free | Free | — |
Alternatives to Keystroke
View alln8n
n8n is a visual workflow automation platform where AI agents and integrations run on a canvas you can read, trace, and audit.
Openbot
OpenBot is a chief-of-staff AI agent that plans your goals and puts Claude Code, Codex, Cursor and other specialists to work on a managed
AutoGen Studio
Open-source web UI for prototyping multi-agent AI teams on Microsoft's AutoGen AgentChat without writing orchestration code.
Tools that pair well with Keystroke
Common stack mates teams adopt alongside Keystroke, with the specific reason each pairing earns its keep.
n8n
n8n is a visual workflow automation platform where AI agents and integrations run on a canvas you can read, trace, and audit.
Openbot
OpenBot is a chief-of-staff AI agent that plans your goals and puts Claude Code, Codex, Cursor and other specialists to work on a managed
AutoGen Studio
Open-source web UI for prototyping multi-agent AI teams on Microsoft's AutoGen AgentChat without writing orchestration code.
Keystroke vs its closest rival
Keystroke's closest head-to-head is with Temporal Ai — If you need to rapidly build AI workflows with a visual editor and connect to common SaaS tools, Keystroke is your pick. But if your agents must survive crashes, retries, and long-running processes, Temporal's durable execution is what you truly need for production reliability. Verdict generated
Head-to-head comparisons
Keystroke vs Temporal Ai
If you need to rapidly build AI workflows with a visual editor and connect to common SaaS tools, Keystroke is your pick. But if your agents must survive crashes, retries, and long-running processes, Temporal's durable execution is what you truly need for production reliability.
GeneratedKeystroke vs Air Ai
If you're a developer building custom AI workflows, Keystroke is the agile, cost-effective choice—it's freemium and integrates with the AI models you already use. But if you're in defense or government and need to compress materiel release timelines and achieve mission-critical readiness, Air AI is the specialized powerhouse—recently winning a $31M contract and expanding its Pittsburgh office, it's clearly the defense-first platform. Pick based on your domain, not just features.
GeneratedKeystroke vs Cryptohopper
If you're a developer building custom AI agents and want flexibility to orchestrate models and APIs, Keystroke is your platform. If you're a crypto trader or a developer who needs live market data feed to AI agents, Cryptohopper is the clear choice — its MCP server and recent guides show a strong focus on AI-integrated trading. They're not direct competitors; pick based on your domain.
GeneratedFrequently asked questions
Used Keystroke? Help shape our editorial sentiment research.