K
Kit by Speakeasy
A coding agent runtime that gives one composable tool to write and run programs, reducing model round trips for faster, cheaper agentic coding.
MIT licensedRust single static binaryACP v1/v2 + A2A + MCP supportBuilt by SpeakeasyDocker images available
Opportunity score
58/100
A technically interesting open-source runtime in a hot market, but monetization and distribution are unproven and competition is extremely intense.
Founder verdict
MAYBE
A technically promising, open-source agent runtime in an attractive market that is too early and too crowded to call a clear business.
What is Kit by Speakeasy?
Kit by Speakeasy is an open-source coding agent runtime built in Rust and distributed as a single static binary. It differentiates from Claude Code, Codex CLI, and Cursor by exposing only one main tool to the model—compose—which accepts a short Runlet program that can read files, run tests, apply edits, retry commands, delegate to subagents, and return structured data in a single model round trip. Kit supports ACP v1/v2, A2A, MCP, Agent Skills, and Agent Plugin packages, uses append-only JSONL transcripts for crash-safe long sessions, and supports models via ChatGPT OAuth, OpenRouter, and Speakeasy AI Control Plane. The project is early: 27 stars, 3 forks, no public revenue, growth, pricing, or customer metrics. The business model appears to be open-source adoption with possible monetization through Speakeasy's AI Control Plane, but that is not stated publicly.
Get the full Kit by Speakeasy playbook
The complete reverse-engineering — what to copy, what to avoid, and exactly what to build instead.
- Full opportunity score across 8 dimensions
- The real problem & why customers pay
- Why it's winning — with evidence
- Complete business reverse-engineering
- Competitive advantages & moat analysis
- Weaknesses, risks & what to avoid
- Clone strategy — what to build instead
- Week-by-week MVP roadmap
- Recommended technical stack
- Founder verdict & highest-leverage move