Skip to content

dl4rce/flaiwheel

v3.9.13 Feature

This release adds 2 notable features for engineering teams evaluating rollout.

Published 3mo MCP Search & Web
✓ No known CVEs patched
Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Summary

AI summary

Two-pass classifier improves cold-start classification quality with path heuristics and code‑specific embedding templates.

Full changelog

Improved: Cold-Start Classification Quality

Two-pass classifier in code_analyzer.py — no new dependencies, no Docker rebuild.

Pass 1 — Path heuristics (zero model cost)

High-confidence pattern matching on filename/path runs first:

| Pattern | Category |
|---|---|
| supabase/functions/*/index.ts | api |
| test_*, *.spec.*, */tests/* | tests |
| config, settings, docker, deploy | setup |
| utils, helpers, shared, lib | best-practices |
| error, retry, fallback | bugfix-log |
| main, server, router, handler, api, middleware | api |

Resolves ~60-70% of files with 0.9 confidence — no embedding call needed.

Pass 2 — Embedding fallback (code-specific templates)

For remaining files, uses new _CODE_CATEGORY_TEMPLATES tuned to what source code looks like (route handlers, assert statements, env vars) instead of the doc-oriented templates that caused changelog over-classification on large mixed codebases.

Upgrade

curl -sSL https://raw.githubusercontent.com/dl4rce/flaiwheel/main/scripts/install.sh | bash

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Track dl4rce/flaiwheel

Get notified when new releases ship.

Sign up free

About dl4rce/flaiwheel

Self-hosted memory and governance layer for AI coding agents. 28 MCP tools with structured knowledge capture, hybrid search (semantic + BM25 + cross-encoder reranking), behavioral documentation nudges, cold-start codebase analyzer, and git-native storage. Single Docker container, zero cloud dependencies.

All releases →

Beta — feedback welcome: [email protected]