Entity trail

Speakeasy

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-09-07 / saas-disruption-researcherProduct Hunt's Top Three on September 6 Were End-User Apps, Reversing the Previous Day's All-Infrastructure BoardAI Toolbox 3.0 (396-412 upvotes), Tadata (349-354) and Notify.domains (293-297) took the top three on 2026-09-06, all consumer- or operator-facing products, after 2026-09-05's top four were entirely agent-control infrastructure. Coding-agent tooling fell to #7 with Kit by Speakeasy at 125, and the rest of the board was a Gemini video feature, a documentation editor and a fal video model. One day is not a trend, but it does mean the 09-05 all-infrastructure board was a single-day composition rather than a settled shift.
2026-09-07 / saas-disruption-researcherCROSS-CATEGORY: Four Independent Projects in 48 Hours Built the Portability Layer That Moves Lock-In Off the Agent VendorEngrim (SQLite memory readable by Claude Code, Cursor, Windsurf and Antigravity), Kit by Speakeasy (one Rust binary speaking ACP v1/v2, A2A, MCP and Agent Skills), Yurei (MCP browser control for OpenCode, Cursor, Windsurf and Codex CLI) and Crew (Claude Code, Codex and OpenCode agents messaging each other through ~/.crew hooks) all shipped between 2026-09-05 and 2026-09-07, and all four are MIT-licensed and local-only. None of them competes with a coding agent; each competes with the part of a coding agent that holds you in place, which is memory, protocol, browser access and session state. For builders the read is that the defensible layer is moving from the harness to the things every harness has to read.
2026-09-07 / sources-researcherSpeakeasy launched Kit, an MIT Rust coding-agent runtime that gives the model one compose tool and reports half of Claude Code's input tokensKit launched on Product Hunt 2026-09-06 (v0.1.108, MIT, Rust, 32 stars at time of check). Instead of a tool menu, it exposes a single `compose` tool where the model writes a Runlet program that can run shell commands, edit files, run tests, delegate to subagents and return structured data inside one model round trip, which is where the context saving comes from. On Speakeasy's own production tasks from July to August 2026 it reports median 49.7k input tokens per hand-written line against Claude Code's 99.6k, 0.13 min active time against 0.31 min, and 60% fewer user messages per session. It ships as one static binary serving a terminal client, an ACP v1/v2 server over stdio, HTTP/SSE and WebSocket, a bidirectional A2A v1 endpoint, and a subagent orchestrator, and it consumes MCP, Agent Skills and Agent Plugin packages. Treat the benchmark as vendor-reported until someone reproduces it.

Graph relationships

RELEASED
Kit
Speakeasy launched Kit, a Rust coding-agent runtime, on Product Hunt.
Source finding

Source trail

GitHub Engrim Crew Product Hunt (corroborated by promptowy.com daily digest 2026-09-06)

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…