pi-herdsman 0.5.2 → 0.6.1
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 -5
- package/dist/agent-definitions/generalist.md +0 -2
- package/dist/agent-definitions/implementer.md +0 -2
- package/dist/agent-definitions/researcher.md +0 -2
- package/dist/agent-definitions/reviewer.md +0 -2
- package/dist/agent-definitions/scout.md +0 -2
- package/dist/index.js +75 -68
- package/docs/agent-api.md +1 -1
- package/docs/concepts/agents.md +9 -6
- package/docs/concepts/lifecycle.md +4 -2
- package/docs/development/smoke-testing.md +7 -5
- package/docs/development/validation.md +17 -12
- package/docs/getting-started.md +15 -12
- package/docs/guides/agent-definitions.md +8 -2
- package/docs/guides/customizing-agents.md +12 -11
- package/docs/guides/handoffs.md +7 -1
- package/docs/guides/recovery.md +3 -1
- package/docs/reference/agent-definition-schema.md +50 -25
- package/docs/reference/agent.md +2 -1
- package/docs/reference/commands.md +14 -10
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -2,6 +2,11 @@
|
|
|
2
2
|
|
|
3
3
|
|
|
4
4
|
|
|
5
|
+
[](https://www.npmjs.com/package/pi-herdsman)
|
|
6
|
+
[](https://github.com/boadij/pi-herdsman/actions/workflows/validate.yml)
|
|
7
|
+
[](https://github.com/boadij/pi-herdsman/actions/workflows/validate.yml)
|
|
8
|
+
[](LICENSE)
|
|
9
|
+
|
|
5
10
|
**Asynchronous [Pi](https://github.com/earendil-works/pi) subagents and agent fleet orchestration with nested delegation, parallel execution, and supervision in [herdr](https://github.com/herdrdev/herdr).**
|
|
6
11
|
|
|
7
12
|
Keep the conversation. Delegate the work.
|
|
@@ -47,20 +52,29 @@ chief
|
|
|
47
52
|
|
|
48
53
|
```sh
|
|
49
54
|
pi install npm:pi-herdsman
|
|
55
|
+
herdr integration install pi
|
|
56
|
+
```
|
|
57
|
+
|
|
58
|
+
Start Herdr in your project:
|
|
59
|
+
|
|
60
|
+
```sh
|
|
61
|
+
herdr
|
|
50
62
|
```
|
|
51
63
|
|
|
52
|
-
|
|
64
|
+
Then run Pi in the Herdr pane:
|
|
53
65
|
|
|
54
|
-
|
|
66
|
+
```sh
|
|
67
|
+
pi
|
|
68
|
+
```
|
|
55
69
|
|
|
56
70
|
Ask Pi normally:
|
|
57
71
|
|
|
58
72
|
```text
|
|
59
|
-
Use
|
|
73
|
+
Use scout to inspect this repository.
|
|
60
74
|
```
|
|
61
75
|
|
|
62
|
-
That
|
|
63
|
-
|
|
76
|
+
That's enough. The agent runs asynchronously while the lead conversation remains
|
|
77
|
+
available.
|
|
64
78
|
|
|
65
79
|
Other useful requests look the same:
|
|
66
80
|
|
|
@@ -1,8 +1,6 @@
|
|
|
1
1
|
---
|
|
2
2
|
name: generalist
|
|
3
3
|
description: General-purpose execution agent for scoped tasks that do not fit scout, researcher, implementer, or reviewer
|
|
4
|
-
model: openai-codex/gpt-5.6-luna
|
|
5
|
-
thinking: medium
|
|
6
4
|
agents: ["scout", "researcher"]
|
|
7
5
|
systemPromptMode: replace
|
|
8
6
|
inheritProjectContext: true
|
|
@@ -1,8 +1,6 @@
|
|
|
1
1
|
---
|
|
2
2
|
name: implementer
|
|
3
3
|
description: Focused implementation agent for a resolved change; use when the required behavior is already decided and the task is to edit, test, and report
|
|
4
|
-
model: openai-codex/gpt-5.6-luna
|
|
5
|
-
thinking: high
|
|
6
4
|
agents: ["scout"]
|
|
7
5
|
systemPromptMode: replace
|
|
8
6
|
inheritProjectContext: true
|
|
@@ -1,8 +1,6 @@
|
|
|
1
1
|
---
|
|
2
2
|
name: researcher
|
|
3
3
|
description: External research specialist for questions that require web, documentation, standards, vendor, or other authoritative evidence beyond the repository; use for current facts, API behavior, comparisons, and source-backed recommendations
|
|
4
|
-
model: openai-codex/gpt-5.6-luna
|
|
5
|
-
thinking: medium
|
|
6
4
|
systemPromptMode: replace
|
|
7
5
|
inheritProjectContext: true
|
|
8
6
|
inheritGlobalContext: false
|
|
@@ -1,8 +1,6 @@
|
|
|
1
1
|
---
|
|
2
2
|
name: reviewer
|
|
3
3
|
description: Independent read-only reviewer for plans, diffs, implementations, and codebase health; use when work needs verification, missing-case analysis, or regression review rather than modification
|
|
4
|
-
model: openai-codex/gpt-5.6-luna
|
|
5
|
-
thinking: medium
|
|
6
4
|
agents: ["scout", "researcher"]
|
|
7
5
|
systemPromptMode: replace
|
|
8
6
|
inheritProjectContext: true
|
|
@@ -1,8 +1,6 @@
|
|
|
1
1
|
---
|
|
2
2
|
name: scout
|
|
3
3
|
description: Read-only codebase reconnaissance for unfamiliar areas; use to find entry points, trace flows, dependencies, constraints, and risks before deciding or editing
|
|
4
|
-
model: openai-codex/gpt-5.6-luna
|
|
5
|
-
thinking: medium
|
|
6
4
|
systemPromptMode: replace
|
|
7
5
|
inheritProjectContext: true
|
|
8
6
|
inheritGlobalContext: false
|