You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: CONTRIBUTING.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -20,7 +20,7 @@ has its own scope:
20
20
|[config](https://github.com/lpb-stack/config)| Pi settings, MCP servers, custom skills, subagents presets |
21
21
|[pi](https://github.com/earendil-works/pi)| Upstream of the coding agent core — installed from npm at `LPB_PI_VERSION`, not a stack repo since the 2026-08-31 de-fork; patches go upstream |
22
22
|[lemonade-pi-plugin](https://github.com/lpb-stack/lemonade-pi-plugin)| Qwen thinking + vision support (fork) |
23
-
|[pi-subagents](https://github.com/lpb-stack/pi-subagents)|Centralized subagent model registry (fork)|
23
+
|[pi-subagents](https://github.com/tintinweb/pi-subagents)|Upstream of the subagent model registry — installed from npm, not a stack repo since the 2026-09-10 de-fork |
Copy file name to clipboardExpand all lines: README.md
+28-7Lines changed: 28 additions & 7 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -6,6 +6,26 @@ optimized for Qwen models served locally through **Lemonade**, targeting
6
6
AMD Strix Halo hardware. No cloud LLM required; your code and data stay
7
7
on your machine.
8
8
9
+
## What You Get
10
+
11
+
Everything below ships in the base image — no extra setup beyond the
12
+
[Quick Start](#quick-start).
13
+
14
+
| Capability | What it means in practice |
15
+
|---|---|
16
+
|**Local inference, zero cloud**| All model calls go to the local Lemonade server (GGUF models, AMD Strix Halo optimized). Your code, data, and prompts never leave the machine. |
|**Lemonade** (host prerequisite) | Local model server — AMD/Strix Halo-optimized, serves GGUF models; on this host or any reachable host; model chosen by the user at setup (no default) |
@@ -199,9 +219,10 @@ flowchart TB
199
219
200
220
## Forked Repos & Upstream Policy
201
221
202
-
Pi itself is no longer a fork: the image installs mainstream pi from the
203
-
npm registry at `LPB_PI_VERSION` (the `lpb-stack/pi` fork is retired — see
204
-
[Fork improvements](doc/fork-improvements.md) for the retirement note).
222
+
Pi itself is not a repo: the image installs mainstream pi from the npm
223
+
registry at `LPB_PI_VERSION`. The only fork is `lemonade-pi-plugin`; retired
|**lemonade-pi-plugin**|`lemonade-sdk/lemonade-pi-plugin` (no stable release) | Qwen thinking + vision support | follow upstream `main`, check periodically |
214
-
|**pi-subagents**|`tintinweb/pi-subagents` (v0.16.1) |centralized local-first subagent model registry |follow upstream; merge + repair as needed|
235
+
|**pi-subagents**|`tintinweb/pi-subagents` (upstream, npm) |installed from upstream npm — no local fork; subagent model registry |track upstream releases|
215
236
|**lpb-memory**|*(independent project)*| Pi memory extension (subprocess reviews) | no upstream to track |
216
237
|**config** / **devstack**| — | own | own |
217
238
@@ -238,7 +259,7 @@ GitHub Actions (`.github/workflows/build-and-publish.yml`) runs on:
238
259
Versioning is **manual**: `lpb-devstack bump` commits a new `VERSION`, and CI
239
260
builds + tags only when VERSION changed in the pushed commit. Pipeline jobs:
-[lpb-stack/pi-subagents](https://github.com/lpb-stack/pi-subagents) — subagent model registry
339
+
-[tintinweb/pi-subagents](https://github.com/tintinweb/pi-subagents) — subagent model registry (upstream, installed via npm; the `lpb-stack/pi-subagents` fork is retired)
0 commit comments