claude-smart 0.1.8 → 0.1.9
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +19 -23
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -6,14 +6,14 @@
|
|
|
6
6
|
claude-smart
|
|
7
7
|
</h1>
|
|
8
8
|
|
|
9
|
-
<h4 align="center">The <a href="https://claude.com/claude-code" target="_blank">Claude Code</a> plugin that makes Claude Code self-improve as you use it —
|
|
9
|
+
<h4 align="center">The <a href="https://claude.com/claude-code" target="_blank">Claude Code</a> plugin that makes Claude Code self-improve as you use it — so Claude Code stops repeating mistakes and gets better every session.</h4>
|
|
10
10
|
|
|
11
11
|
<p align="center">
|
|
12
12
|
<a href="LICENSE">
|
|
13
13
|
<img src="https://img.shields.io/badge/License-Apache%202.0-blue.svg" alt="License">
|
|
14
14
|
</a>
|
|
15
15
|
<a href="plugin/pyproject.toml">
|
|
16
|
-
<img src="https://img.shields.io/badge/version-0.1.
|
|
16
|
+
<img src="https://img.shields.io/badge/version-0.1.9-green.svg" alt="Version">
|
|
17
17
|
</a>
|
|
18
18
|
<a href="plugin/pyproject.toml">
|
|
19
19
|
<img src="https://img.shields.io/badge/python-%3E%3D3.12-brightgreen.svg" alt="Python">
|
|
@@ -37,12 +37,15 @@
|
|
|
37
37
|
</p>
|
|
38
38
|
|
|
39
39
|
<p align="center">
|
|
40
|
-
It learns both corrections and successful execution patterns—so Claude Code avoids repeating mistakes and reuses what works.
|
|
40
|
+
It learns both corrections and successful execution patterns—so Claude Code avoids repeating mistakes and reuses what works. Claude Code steadily adapts to <i>how you like</i> to work—across projects, codebases, and sessions.
|
|
41
41
|
</p>
|
|
42
42
|
|
|
43
43
|
<p align="center">
|
|
44
|
-
<b>Head-to-head vs <code>claude-mem</code
|
|
44
|
+
<b>Head-to-head vs <code>claude-mem</code>.</b> Evaluated across 12 scripted scenarios.
|
|
45
|
+
<br>
|
|
46
|
+
<b>~2.7× higher overall quality </b> · Right user correction learnt every time vs none · See <a href="benchmarks/memory_comparison/EXPERIMENT.md">EXPERIMENT.md</a> for details.
|
|
45
47
|
</p>
|
|
48
|
+
|
|
46
49
|
---
|
|
47
50
|
|
|
48
51
|
## Why Learning, Not Memory
|
|
@@ -53,29 +56,22 @@ Most memory solutions are still mostly informative—Claude remembers what happe
|
|
|
53
56
|
|
|
54
57
|
Four ways this changes what Claude Code can do for you:
|
|
55
58
|
|
|
56
|
-
- **
|
|
59
|
+
- 💡 **Stop repeating the same mistakes:** Produces actionable playbooks Claude can follow next time; memory only records what happened.
|
|
57
60
|
|
|
58
|
-
> *Example:* you tell Claude to stop running `npm test` without `--run` because watch mode hangs
|
|
59
|
-
> **Memory:** “user was annoyed about npm test hanging
|
|
61
|
+
> *Example:* you tell Claude to stop running `npm test` without `--run` because watch mode hangs.<br>
|
|
62
|
+
> **Memory:** “user was annoyed about npm test hanging”<br>
|
|
60
63
|
> **Learning:** “always pass `--run` to `npm test` in this repo — default watch mode blocks CI”
|
|
61
64
|
|
|
62
|
-
- **
|
|
63
|
-
> *Example:* Claude spends several iterations trying to start the local dev environment before discovering that this repo requires `pnpm dev:all` instead of the usual `npm run dev
|
|
64
|
-
> **Memory:** “user mentioned that `npm run dev` did not work
|
|
65
|
-
> **Learning:** “for this repo, always use `pnpm dev:all` to start the full local stack — `npm run dev` only starts the frontend and causes missing service errors”
|
|
66
|
-
|
|
67
|
-
Instead of re-exploring the same setup problem next time, Claude starts from the proven path—reducing planning steps, latency, and token usage.
|
|
68
|
-
|
|
69
|
-
- **Project-wide, not session-siloed:** Session memory disappears with the conversation. The project playbook persists and improves across every session in that repo.
|
|
70
|
-
|
|
71
|
-
- **Compact:** Distilled, deduplicated playbooks stay in dozens of tokens—not thousands—even as the project grows.
|
|
65
|
+
- 🚀 **Start from the optimized path:** Preserves and optimizes execution paths so Claude can reuse what already works.
|
|
66
|
+
> *Example:* Claude spends several iterations trying to start the local dev environment before discovering that this repo requires `pnpm dev:all` instead of the usual `npm run dev`.<br>
|
|
67
|
+
> **Memory:** “user mentioned that `npm run dev` did not work”<br>
|
|
68
|
+
> **Learning:** “for this repo, always use `pnpm dev:all` to start the full local stack — `npm run dev` only starts the frontend and causes missing service errors”
|
|
72
69
|
|
|
73
|
-
|
|
70
|
+
Instead of re-exploring, Claude starts from the proven path—reducing planning steps, latency, and token usage.
|
|
74
71
|
|
|
75
|
-
-
|
|
76
|
-
- **Project Playbook** → durable rules and optimized execution paths for how Claude should behave
|
|
72
|
+
- 🌐 **Project-wide, not session-siloed:** Session memory disappears with the conversation. The project playbook persists and improves across every session in that repo.
|
|
77
73
|
|
|
78
|
-
|
|
74
|
+
- 🪶 **Better context without prompt bloat:** Distilled, deduplicated playbooks stay in dozens of tokens—not thousands—even as the project grows.
|
|
79
75
|
|
|
80
76
|
---
|
|
81
77
|
|
|
@@ -118,7 +114,6 @@ Developing the plugin itself? See [DEVELOPER.md](./DEVELOPER.md#developing-local
|
|
|
118
114
|
- ⚡ **Fully automatic learning** — Every user turn, tool call, and assistant response is captured via lifecycle hooks and extracted into rules without you running anything.
|
|
119
115
|
- 📈 **Updates with every session** — Playbooks auto-merge, supersede, and archive as your project evolves — the playbook sharpens with use instead of bloating.
|
|
120
116
|
> *e.g.* you correct the same `npm test --run` gotcha twice → **claude-smart** consolidates them into one stronger rule. Later you switch the policy to `pnpm test` → the old rule is archived and the new one supersedes it, no manual cleanup.
|
|
121
|
-
- 🎯 **Two-tier scope** — Per-session profiles for the current conversation; cross-session playbooks for the whole project.
|
|
122
117
|
- 🔌 **No external API call** — semantic search runs on an in-process ONNX embedder (all-MiniLM-L6-v2), and all data (profiles, playbooks, interaction buffers) is stored locally on your machine (`~/.reflexio/` and `~/.claude-smart/`).
|
|
123
118
|
- 🔎 **Hybrid search** — Playbooks and profiles are indexed with vector + BM25 search for fast, robust retrieval.
|
|
124
119
|
- 🧪 **Offline resilience** — If the reflexio backend is down, hooks buffer to disk; the next successful publish drains them.
|
|
@@ -128,7 +123,7 @@ Developing the plugin itself? See [DEVELOPER.md](./DEVELOPER.md#developing-local
|
|
|
128
123
|
|
|
129
124
|
## Dashboard
|
|
130
125
|
|
|
131
|
-
A
|
|
126
|
+
A web UI for browsing session histories, inspecting user profiles, and editing project playbooks. The dashboard auto-starts alongside the backend, so you can open **http://localhost:3001** directly. Or run `/claude-smart:dashboard` in Claude Code to launch dashboard in browser.
|
|
132
127
|
|
|
133
128
|
<p align="center">
|
|
134
129
|
<img src="assets/profile_dashboard.png" alt="Profile dashboard" width="49%">
|
|
@@ -156,6 +151,7 @@ See [ARCHITECTURE.md](./ARCHITECTURE.md) for hooks, data flow, and reflexio deta
|
|
|
156
151
|
|
|
157
152
|
| Command | What it does |
|
|
158
153
|
| --- | --- |
|
|
154
|
+
| `/dashboard` | Open the dashboard in your browser, auto-starting the reflexio backend and dashboard services if they aren't already running. |
|
|
159
155
|
| `/show` | Print the current project playbook plus the current session's user profiles (same markdown that `SessionStart` injects). Use it to audit what playbooks and preferences Claude is being told to follow. |
|
|
160
156
|
| `/learn` | Force reflexio to run extraction *now* on the current session's unpublished interactions. Without this, extraction runs at the end of the session or on reflexio's batch interval. |
|
|
161
157
|
| `/tag [note]` | Tag the most recent turn as a correction, for cases the automatic heuristic missed. The note becomes the correction description the extractor sees. |
|