talking-stick 0.1.0-alpha.6 → 0.1.0
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 +2 -2
- package/dist/mcp-server.js +1 -1
- package/docs/releases/0.1.0.md +37 -0
- package/package.json +1 -1
- package/skills/talking-stick/SKILL.md +2 -2
package/README.md
CHANGED
|
@@ -2,7 +2,7 @@
|
|
|
2
2
|
|
|
3
3
|
An MCP coordination server that lets multiple AI coding agents share a single workspace without stepping on each other. One agent holds the stick at a time; handoffs carry structured context so the next agent doesn't have to re-derive it.
|
|
4
4
|
|
|
5
|
-
**Version:** 0.1.0
|
|
5
|
+
**Version:** 0.1.0. Multi-process-safe (SQLite WAL), liveness-aware, no daemon. Supports Claude Code, Codex CLI, Gemini CLI, and OpenCode out of the box.
|
|
6
6
|
|
|
7
7
|
## Quickstart
|
|
8
8
|
|
|
@@ -27,7 +27,7 @@ That's it. The next time two agents `cd` into the same repo, they see each other
|
|
|
27
27
|
|
|
28
28
|
| Method | Command | Notes |
|
|
29
29
|
|---|---|---|
|
|
30
|
-
| **From npm** | `npm i -g talking-stick` | Published as `0.1.0
|
|
30
|
+
| **From npm** | `npm i -g talking-stick` | Published as `0.1.0`. Requires Node ≥ 22. |
|
|
31
31
|
| **From GitHub** | `npm i -g github:mostlydev/talking-stick` | Tracks the `master` branch; builds on install via the `prepare` hook. |
|
|
32
32
|
| **From source** | `git clone … && npm install && npm link` | For contributors. |
|
|
33
33
|
|
package/dist/mcp-server.js
CHANGED
|
@@ -27,7 +27,7 @@ export function createMcpServer(service = new TalkingStickService()) {
|
|
|
27
27
|
const resolveConnectionIdentity = createConnectionIdentityResolver();
|
|
28
28
|
const server = new McpServer({
|
|
29
29
|
name: "talking-stick",
|
|
30
|
-
version: "0.1.0
|
|
30
|
+
version: "0.1.0"
|
|
31
31
|
});
|
|
32
32
|
server.registerTool("list_rooms", {
|
|
33
33
|
title: "List Rooms",
|
|
@@ -0,0 +1,37 @@
|
|
|
1
|
+
# Talking Stick 0.1.0
|
|
2
|
+
|
|
3
|
+
Date: 2026-04-27
|
|
4
|
+
|
|
5
|
+
First non-alpha release. This promotes the current single-host coordination
|
|
6
|
+
surface from alpha to the initial SemVer baseline.
|
|
7
|
+
|
|
8
|
+
## What is in the baseline
|
|
9
|
+
|
|
10
|
+
- MCP room coordination: `join_path`, `wait_for_turn`, `heartbeat`,
|
|
11
|
+
`release_stick`, `pass_stick`, `takeover_stick`, state reads, event reads,
|
|
12
|
+
and non-owner notes.
|
|
13
|
+
- Human `tt` CLI flows for joining, waiting, releasing, passing, takeover,
|
|
14
|
+
notes, room inspection, installers, skill installers, and self-update.
|
|
15
|
+
- SQLite WAL-backed persistence with multi-process safety, lease fencing,
|
|
16
|
+
liveness-aware recovery, fair release selection, and ephemeral room cleanup.
|
|
17
|
+
- User-global MCP and skill install support for Claude Code, Codex CLI, Gemini
|
|
18
|
+
CLI, and OpenCode.
|
|
19
|
+
|
|
20
|
+
## Changed
|
|
21
|
+
|
|
22
|
+
### Non-alpha package version
|
|
23
|
+
|
|
24
|
+
Package metadata, README version text, and the MCP server handshake now advertise
|
|
25
|
+
`0.1.0`.
|
|
26
|
+
|
|
27
|
+
### Long-poll skill guidance
|
|
28
|
+
|
|
29
|
+
The bundled skill now uses `110000` ms as the default client-safe
|
|
30
|
+
`wait_for_turn` long-poll budget.
|
|
31
|
+
|
|
32
|
+
## Verification
|
|
33
|
+
|
|
34
|
+
- `npm run typecheck`
|
|
35
|
+
- `npm test` — 208 tests across 14 files
|
|
36
|
+
- `git diff --check`
|
|
37
|
+
- `npm pack --dry-run --ignore-scripts`
|
package/package.json
CHANGED
|
@@ -53,11 +53,11 @@ Keep the wait input minimal:
|
|
|
53
53
|
```json
|
|
54
54
|
{
|
|
55
55
|
"room_id": "<room_id from join_path>",
|
|
56
|
-
"max_wait_ms":
|
|
56
|
+
"max_wait_ms": 110000
|
|
57
57
|
}
|
|
58
58
|
```
|
|
59
59
|
|
|
60
|
-
`max_wait_ms` is optional. Use the longest client-safe wait you can support:
|
|
60
|
+
`max_wait_ms` is optional. Use the longest client-safe wait you can support: 110000 ms is a good MCP default when the harness can tolerate it; 180000 ms is fine only when the tool/client timeout is known to exceed that. If the call times out at the harness layer, fall back to a shorter value and call again. Do not send `cursor`, even if an old tool schema still exposes it; `wait_for_turn` is cursor-free, and resumable event replay belongs to `get_room_events`.
|
|
61
61
|
|
|
62
62
|
Possible outcomes:
|
|
63
63
|
|