GizClaw/flowcraft
Frameworks & Eval · tested 2026-09-30 · by the Hlido desk, not the vendor
In short: Actively maintained Go SDK for stateful, voice-capable agents — clean repo hygiene, but no drop-in agent-consumable surface documented.
9 PASS · 1 FAIL of 10 public-surface claims
Quick answer
GizClaw/flowcraft scores 82/100 (STEADY) on Hlido’s independent, hands-on test (reviewed 2026-09-30). STEADY (82) because the repository is healthy on every hygiene axis we measure — permissive licence, active maintenance, releases, CI, real star traction — which is what the repo-surface track rewards. Pricing: Open source (free entry point documented).
flowcraft presents itself as a production-grade Go SDK for building agents with long-term memory, knowledge retrieval and a voice pipeline, runnable as a library, a daemon or a real-time stream. On the repo signals we can measure it holds up well: MIT-licensed, 486 stars, a tagged release, CI workflows present, and a push the same day we captured it — this is a live project, not an abandoned experiment. Go is a sensible choice for a long-running agent daemon where concurrency and a single static binary matter. The one gap the engine flagged is real: nothing in the repository documents an agent-consumable surface (no MCP config, no CLI-install recipe surfaced), so another agent can't yet drive flowcraft without a developer wiring the SDK in first. That keeps it a library you build with, not a component an agent picks up autonomously. For a Go team that wants memory and voice primitives under one roof, it is a credible starting point; treat the 'voice' framing as one capability among several rather than the headline.
Why STEADY
STEADY (82) because the repository is healthy on every hygiene axis we measure — permissive licence, active maintenance, releases, CI, real star traction — which is what the repo-surface track rewards. Held below VITAL by the repo-surface ceiling (no hands-on run) and by the missing agent-consumable documentation, the single failed checklist item.
Public-surface checklist
- PASS Repo reachable (required) — GH API 200 for GizClaw/flowcraft
- PASS Readme present (required) — README length 15367
- PASS License present (required) — MIT
- PASS Install documented (required) — install/usage section found in README
- PASS Active 12mo (required) — last push 0d ago
- PASS Releases present — latest vesseld/v0.1.0
- PASS Community traction — 486 stars
- PASS Ci or tests — 12 workflow file(s)
- PASS Recent commit 90d — last push 0d ago
- FAIL Agent consumable — none found
What we saw
1 screenshot captured by the Hlido engine during the reviewed run (run-bcc4cfb34a31d83c-github-com). Our own captures — not vendor marketing material.
What it does well
- MIT-licensed with a tagged release — clear terms and a versioned artefact to pin
- Actively maintained: last push the same day as capture, CI workflows in place
- Real community traction (486 stars) for a niche infrastructure library
- Go gives a single static binary and solid concurrency for a long-running agent daemon
What it fails at
- No agent-consumable surface documented (no MCP config or CLI-install recipe found) — a developer must wire the SDK in by hand
- Not hands-on tested — memory, retrieval and voice quality are unverified beyond the README
- Voice is marketed prominently but is one capability among several, not a proven differentiator
Best for
- Go teams that want memory + retrieval + voice primitives in one library
- Developers comfortable integrating an SDK rather than consuming a ready endpoint
- Projects that need a long-running agent daemon as a static binary
Not recommended for
- Agents that need to self-integrate without a developer wiring the SDK
- Teams outside the Go ecosystem
- Buyers wanting a hands-on-verified voice product rather than a framework
Pricing & access
- ModelOpen source
- Free entry pointYes — a free tier or open-source edition is documented
Derived from Hlido-held evidence only (engine checklist + editorial text); quotes are verbatim from the scorecard; not vendor-supplied; re-derived daily. Verify current prices on the vendor's pricing page. Last verified 2026-07-16.
Related agents
Agent relevance
SDK
Agentic-Commerce Readiness 37/100 · SURFACE-ONLY
Independent readiness for agent delegation & transaction. How it’s scored · check live
Imported as a Go SDK / library, or run as a daemon. No MCP or documented CLI surface found, so agent consumption requires a developer to build the integration.
Agent-friendly score: 5/10
Score over time
The longitudinal record — every point is the score as published on that date. Raw series.
