@yemi33/minions 0.1.235 → 0.1.236
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/CHANGELOG.md +4 -1
- package/README.md +5 -9
- package/docs/auto-discovery.md +1 -1
- package/docs/command-center.md +2 -2
- package/docs/deprecated.json +10 -1
- package/docs/distribution.md +0 -1
- package/package.json +1 -1
- package/routing.md +0 -1
package/CHANGELOG.md
CHANGED
|
@@ -1,11 +1,14 @@
|
|
|
1
1
|
# Changelog
|
|
2
2
|
|
|
3
|
-
## 0.1.
|
|
3
|
+
## 0.1.236 (2026-04-03)
|
|
4
4
|
|
|
5
5
|
### Fixes
|
|
6
6
|
- CLI audit — kill dashboard on restart, expose kill/complete, update help
|
|
7
7
|
- clean _pendingReason on all status→done transitions, not just updateWorkItemStatus
|
|
8
8
|
|
|
9
|
+
### Other
|
|
10
|
+
- docs: fix outdated references across all documentation
|
|
11
|
+
|
|
9
12
|
## 0.1.233 (2026-04-03)
|
|
10
13
|
|
|
11
14
|
### Fixes
|
package/README.md
CHANGED
|
@@ -354,18 +354,14 @@ When an agent fails (timeout, crash, error), the engine marks the work item as `
|
|
|
354
354
|
|
|
355
355
|
## Auto-Discovery Pipeline
|
|
356
356
|
|
|
357
|
-
The engine discovers work from
|
|
357
|
+
The engine discovers work from 4 sources, in priority order:
|
|
358
358
|
|
|
359
359
|
| Priority | Source | Dispatch Type |
|
|
360
360
|
|----------|--------|---------------|
|
|
361
|
-
| 1 |
|
|
362
|
-
| 2 |
|
|
363
|
-
| 3 |
|
|
364
|
-
| 4 |
|
|
365
|
-
| 5 | PRs needing build/test verification | `test` |
|
|
366
|
-
| 6 | Plan items (`plans/*.json`, approved) | `implement` |
|
|
367
|
-
| 7 | Per-project work items | item's `type` |
|
|
368
|
-
| 8 | Central work items | item's `type` |
|
|
361
|
+
| 1 | Pull requests (changes-requested, human feedback, build failures, pending review) | `fix`, `review`, `test` |
|
|
362
|
+
| 2 | Plan items (`plans/*.json`, approved) | `implement` |
|
|
363
|
+
| 3 | Per-project work items | item's `type` |
|
|
364
|
+
| 4 | Central work items (project-agnostic tasks) | item's `type` |
|
|
369
365
|
|
|
370
366
|
Each item passes through: dedup (checks pending, active, AND recently completed), cooldown, and agent availability gates. See `docs/auto-discovery.md` for the full pipeline.
|
|
371
367
|
|
package/docs/auto-discovery.md
CHANGED
|
@@ -20,7 +20,7 @@ tick()
|
|
|
20
20
|
|
|
21
21
|
## Work Discovery
|
|
22
22
|
|
|
23
|
-
`discoverWork()` iterates every project in `config.projects[]` and runs
|
|
23
|
+
`discoverWork()` iterates every project in `config.projects[]` and runs four core discovery sources: pull requests, per-project work items, central work items, and pipeline/scheduled tasks. Results are prioritized: fixes > reviews > implements > work-items.
|
|
24
24
|
|
|
25
25
|
Before scanning, the engine materializes plans and specs into project work items (side-effect writes to `work-items.json`), so they're picked up by the work items source below.
|
|
26
26
|
|
package/docs/command-center.md
CHANGED
|
@@ -13,7 +13,7 @@ CC maintains a true multi-turn session using Claude CLI's `--resume` flag. Unlik
|
|
|
13
13
|
**Session lifecycle:**
|
|
14
14
|
- **Created** on first message (or after expiry/new-session)
|
|
15
15
|
- **Resumed** on subsequent messages via `--resume <sessionId>`
|
|
16
|
-
- **Expires** after 2 hours of inactivity
|
|
16
|
+
- **Expires** after 2 hours of inactivity
|
|
17
17
|
- **Persisted** to `engine/cc-session.json` — survives dashboard restarts
|
|
18
18
|
- **Frontend messages** saved to `localStorage` — survive page refresh
|
|
19
19
|
|
|
@@ -108,7 +108,7 @@ User message (CC panel, doc modal, or steer)
|
|
|
108
108
|
▼
|
|
109
109
|
POST /api/command-center (or /api/doc-chat, /api/steer-document)
|
|
110
110
|
│
|
|
111
|
-
├── Validate session (expiry
|
|
111
|
+
├── Validate session (expiry check)
|
|
112
112
|
├── Build dynamic state preamble (buildCCStatePreamble)
|
|
113
113
|
├── callLLM() with sessionId (resume) or without (new)
|
|
114
114
|
│ model: sonnet
|
package/docs/deprecated.json
CHANGED
|
@@ -1 +1,10 @@
|
|
|
1
|
-
[
|
|
1
|
+
[
|
|
2
|
+
{
|
|
3
|
+
"id": "evaluate-work-type",
|
|
4
|
+
"summary": "evaluate work type removed — eval loop uses review exclusively",
|
|
5
|
+
"deprecated": "2026-04-02",
|
|
6
|
+
"reason": "Consolidated into review + evalLoop config. No separate evaluate dispatch.",
|
|
7
|
+
"locations": ["routing.md: evaluate row removed", "CLAUDE.md: evaluate references removed"],
|
|
8
|
+
"cleanup": "Already cleaned — routing.md and CLAUDE.md updated 2026-04-02"
|
|
9
|
+
}
|
|
10
|
+
]
|
package/docs/distribution.md
CHANGED
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@yemi33/minions",
|
|
3
|
-
"version": "0.1.
|
|
3
|
+
"version": "0.1.236",
|
|
4
4
|
"description": "Multi-agent AI dev team that runs from ~/.minions/ — five autonomous agents share a single engine, dashboard, and knowledge base",
|
|
5
5
|
"bin": {
|
|
6
6
|
"minions": "bin/minions.js"
|
package/routing.md
CHANGED
|
@@ -17,7 +17,6 @@ How the engine decides who handles what. Parsed by engine.js — keep the table
|
|
|
17
17
|
| test | dallas | ralph |
|
|
18
18
|
| ask | ripley | rebecca |
|
|
19
19
|
| verify | dallas | ralph |
|
|
20
|
-
| evaluate | ripley | rebecca |
|
|
21
20
|
| decompose | ripley | rebecca |
|
|
22
21
|
| meeting | ripley | rebecca |
|
|
23
22
|
|