- Home
- Case studies
- Case
AgenKit: AI coding-agent kits, $2.8K in 30 days, seven-harness workflow
AgenKit sells specialist AI coding agents and a software-building workflow for coding harnesses. Its $2.8K in the last 30 days is verified through the payment provider's API by TrustMRR (Polar), as of September 30, 2026.
What it actually sells
AgenKit sells AI coding-agent engineering kits and a workflow for Claude Code, Cursor, Codex, and other harnesses. Its site describes it as “A full team of specialist AI coding agents and the /agenkit workflow for Claude Code, Cursor, and Codex.”
The workflow covers brainstorm, architecture, test-first build, review, and shipping, with approval pauses at each step. It auto-detects the coding harness, installs files into the project, and works across seven harnesses with one kit and license key.
How it charges
AgenKit uses one yearly subscription. The site lists a $3.49/mo founding price for early subscribers, then $15/mo. The first 90 subscribers pay the founding price, and the site reported 85 of 90 founding seats remaining.
The subscription includes updates and new releases while subscribed. The pricing page shows one subscription with the founding price followed by the stated monthly price.
Where customers come from
The site provides documentation from the home page through “Read the docs →”. It also provides CLI installation commands: “$ npm install -g agenkit”.
Installation through npx is supported, with Node.js 18+ required. A seven-harness integration list names Claude Code, Cursor, Codex, OpenCode, Antigravity, GitHub Copilot, and Kimi Code. The founder lists X / Twitter as a growth channel.
What stands out
AgenKit combines specialist AI coding agents, an approval-based workflow, one kit, seven harnesses, and one license key. The kit auto-detects the coding harness and installs the right files when run in a project folder.
The site reports “0+ installs, and counting” and “0+ SaaS products shipped, and counting”. It also states that the kit works in any language and any architecture and receives every update and new release while subscribed.
From the official site
Build any software, at any scale - solo.
One kit, seven harnesses, one license key
Excerpts from the official website, fetched October 1, 2026.
Where it sits in the data
AgenKit's $2.8K in the last 30 days is higher than 85% of the 5,602 projects with payment-verified revenue in our dataset; the median across them is $150. Within the “Dev Tools” category (717 projects) it is higher than 88% (category median $100). Snapshot: 2026-09-02 to 2026-09-30. The dataset only covers founders who chose to publish their numbers, so it leans toward businesses that work; use it for scale, not as an average.
Currency
All figures are in US dollars as published. A rough yuan or euro conversion is not applied here because prices, taxes and acquisition costs differ by country.
Could you build something like this?
My view: demand is clearest with solo software builders and professional senior developers who need a structured workflow across coding-agent harnesses. Distribution would need to reach developers through documentation, command-line installation, npx, integrations, and X / Twitter.
The hard parts are keeping one kit compatible across seven harnesses, supporting approval steps, maintaining updates, and handling the support load that follows a developer-tool subscription. Competition would come from other coding-agent workflows and developer-tool subscriptions.
Verdict: Needs real work.
One thing to take from it
AgenKit’s product shape is a specialist AI coding-agent kit paired with a workflow that moves from brainstorm through shipping while pausing for approval. The same kit supports seven coding harnesses and adapts to existing codebase conventions.
Sources
- AgenKit official website, fetched October 1, 2026
- AgenKit pricing page, fetched October 1, 2026
- TrustMRR: revenue verified through the payment provider's API by TrustMRR (Polar), as of September 30, 2026
Revenue, team and start-cost figures are what the founder reported at the interview date, not current figures and not a prediction of what you will earn. Product and pricing facts come from the business's own website as checked on 30 September 2026. Method & sources · Disclosure