agent-takkub 1.0.44 → 1.0.45
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
CHANGED
|
@@ -33,16 +33,35 @@ npm install -g agent-takkub
|
|
|
33
33
|
|
|
34
34
|
A single AI agent hits a wall on big work: context fills up, sub-tasks collide, and everything runs one-at-a-time. `agent-takkub` runs it like a **real engineering team** — a **Lead** you talk to, and specialist teammates it delegates to, each in its own isolated `claude` process, working **concurrently**.
|
|
35
35
|
|
|
36
|
+
### Orchestration
|
|
37
|
+
|
|
36
38
|
| | |
|
|
37
39
|
| :-- | :-- |
|
|
38
40
|
| 🧠 **Orchestrated teammates** | Converse with the Lead; it spawns, tasks, and manages specialist panes (`frontend`, `backend`, `qa`, `reviewer`, `devops`, `mobile`, …) on demand — only the roles a job actually needs. |
|
|
39
41
|
| 🔀 **True parallelism** | `frontend` and `backend` build a feature at the same time; QA always verifies **last**, against the real running stack. |
|
|
40
42
|
| 🌿 **Branch & worktree isolation** | Parallel teammates each work on their own git branch in an isolated worktree — no commit races, no dirty-state collisions. You merge when ready. |
|
|
41
43
|
| 👥 **Fleet mode** | One toggle scales a role into a fleet (`frontend#1…#K`) sized to your machine — for many independent features or sharded test suites at once. |
|
|
44
|
+
| 🧭 **Plan-first shards** | `takkub assign --role qa --plan --shards N` — a planner splits a big browser-QA sweep into N buckets before fanning out, instead of guessing an even split. |
|
|
42
45
|
| 🖥️ **Steerable, always** | Every pane is a live `claude` shell. Watch output in real time, interrupt, or type straight into any teammate. |
|
|
43
46
|
| 🗂️ **Multi-project tabs** | One isolated Lead per project — no cross-talk. |
|
|
44
47
|
| 🔒 **100% local** | No SaaS middleware. Everything runs on your machine, on your logged-in Claude Code CLI. |
|
|
45
48
|
|
|
49
|
+
### Token efficiency & memory
|
|
50
|
+
|
|
51
|
+
| | |
|
|
52
|
+
| :-- | :-- |
|
|
53
|
+
| 📥 **Pull-on-demand context** | Reference docs, patterns, and CLI cheatsheets live outside the Lead's default prompt — it reads them only when a task needs them, instead of paying for them on every turn. |
|
|
54
|
+
| 🧾 **Per-role memory + L2 archive** | Each teammate keeps a short, token-budgeted "learned notes" file; entries that get trimmed for space are archived (not deleted) to a searchable sibling file instead of being lost. |
|
|
55
|
+
| 🔎 **BM25 search (`takkub search`)** | Full-text search over session logs and role-memory archives, ranked with Okapi BM25 — tokenizes both English words and Thai character trigrams, no external segmenter. |
|
|
56
|
+
| ✅ **Evidence-gated QA/review** | A verify role's "done" report is checked for actual evidence (a file, a test run, an exit code) — reports with no evidence cited get flagged for the Lead instead of trusted at face value. |
|
|
57
|
+
|
|
58
|
+
### Reliability & diagnostics
|
|
59
|
+
|
|
60
|
+
| | |
|
|
61
|
+
| :-- | :-- |
|
|
62
|
+
| 🩺 **`takkub doctor --live`** | Checks the *running* cockpit's spawn queue, not just static config — catches a stuck queue that a config-only check would miss. |
|
|
63
|
+
| ⌨️ **Multiline input that just works** | Shift+Enter / Alt+Enter add a newline instead of submitting, provider-aware (works on claude/gemini panes; skipped where the underlying TUI would misinterpret it) — includes a macOS IME fix for Thai/CJK input dropped after switching input languages. <sub>(community contribution by [@than-aa](https://github.com/than-aa))</sub> |
|
|
64
|
+
|
|
46
65
|
---
|
|
47
66
|
|
|
48
67
|
## 🧠 One team, many model "brains"
|
|
@@ -162,9 +181,11 @@ sequenceDiagram
|
|
|
162
181
|
| `takkub assign --role qa --plan --shards 4 "…"` | Plan-first parallel browser QA (auto fan-out) |
|
|
163
182
|
| `takkub worktree list / merge / clean` | Review + merge isolated branches |
|
|
164
183
|
| `takkub send --to qa "…"` | Message a teammate (Lead CC’d) |
|
|
184
|
+
| `takkub search "…"` | BM25-ranked search over session logs + role-memory archives (Thai + English) |
|
|
165
185
|
| `takkub goal "…"` | Set a session goal injected into every task |
|
|
166
186
|
| `takkub restart` | Restart the whole cockpit from the terminal |
|
|
167
187
|
| `takkub doctor --fix` | Diagnose the environment + auto-repair (add `--install-providers` to also install missing provider CLIs) |
|
|
188
|
+
| `takkub doctor --live` | Same checks, plus a live look at the running cockpit's spawn queue |
|
|
168
189
|
| `takkub provider list` | Show every provider CLI, whether it's installed, and its model |
|
|
169
190
|
| `takkub provider install <name>` | Install one provider CLI (Codex / OpenCode / Kimi) |
|
|
170
191
|
| `takkub provider model <name> [<model>]` | Show or set the model a provider spawns with (`--clear` to reset) |
|
|
Binary file
|
package/package.json
CHANGED