@webpresso/claude-plugin 3.1.8 → 3.1.10
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.
|
@@ -6,7 +6,7 @@
|
|
|
6
6
|
},
|
|
7
7
|
"metadata": {
|
|
8
8
|
"description": "Webpresso agent-kit Claude Code plugin: blueprints, skills, hooks, MCP server",
|
|
9
|
-
"version": "3.1.
|
|
9
|
+
"version": "3.1.10"
|
|
10
10
|
},
|
|
11
11
|
"plugins": [
|
|
12
12
|
{
|
|
@@ -34,5 +34,5 @@
|
|
|
34
34
|
}
|
|
35
35
|
}
|
|
36
36
|
],
|
|
37
|
-
"version": "3.1.
|
|
37
|
+
"version": "3.1.10"
|
|
38
38
|
}
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@webpresso/claude-plugin",
|
|
3
|
-
"version": "3.1.
|
|
3
|
+
"version": "3.1.10",
|
|
4
4
|
"private": false,
|
|
5
5
|
"description": "Claude Code plugin adapter for Webpresso agent-kit skills, commands, and MCP runtime.",
|
|
6
6
|
"homepage": "https://github.com/webpresso/agent-kit#readme",
|
|
@@ -0,0 +1,39 @@
|
|
|
1
|
+
---
|
|
2
|
+
type: skill
|
|
3
|
+
slug: autopilot
|
|
4
|
+
title: Autopilot
|
|
5
|
+
status: active
|
|
6
|
+
scope: repo
|
|
7
|
+
applies_to: [agents]
|
|
8
|
+
related: [goal, plan-refine, verify, testing-philosophy]
|
|
9
|
+
created: "2026-07-10"
|
|
10
|
+
last_reviewed: "2026-07-10"
|
|
11
|
+
name: autopilot
|
|
12
|
+
description: "Drive a brief through the bounded native goal pipeline."
|
|
13
|
+
argument-hint: "<brief|force: brief>"
|
|
14
|
+
---
|
|
15
|
+
|
|
16
|
+
# Autopilot
|
|
17
|
+
|
|
18
|
+
Drive an approved brief through `plan -> execute -> qa -> validate -> land -> complete` using `wp goal run` and the current handoff.
|
|
19
|
+
|
|
20
|
+
## Vague gate
|
|
21
|
+
|
|
22
|
+
A brief is anchored when it includes a file path, symbol, issue number, test name, or numbered steps. Route any unanchored brief through `plan-refine` before execution. A leading `force:` is the only explicit bypass; strip it before creating the goal and record the bypass in the blueprint.
|
|
23
|
+
|
|
24
|
+
## Pipeline
|
|
25
|
+
|
|
26
|
+
1. Plan: refine blueprints and satisfy provenance-backed promotion approvals.
|
|
27
|
+
2. Execute: implement current ready tasks; use only `wp_blueprint_task_verify` to mark them done.
|
|
28
|
+
3. QA: run affected checks and `wp qa`; preserve normalized failure evidence.
|
|
29
|
+
4. Validate: obtain distinct verification and code-review evidence and resolve findings. Each lane appends a tracked `reviews.md` record in one of these exact forms:
|
|
30
|
+
|
|
31
|
+
```text
|
|
32
|
+
<!-- wp:goal-validation {"kind":"verify","reviewer":"<identity>","artifact":"<tracked-path>"} -->
|
|
33
|
+
<!-- wp:goal-validation {"kind":"code-review","reviewer":"<different-identity>","artifact":"<tracked-path>"} -->
|
|
34
|
+
```
|
|
35
|
+
|
|
36
|
+
5. Land: require green PR checks, complete the blueprint in the same PR, merge, and run owner cleanup.
|
|
37
|
+
6. Complete: stop only when `wp goal run` records terminal evidence.
|
|
38
|
+
|
|
39
|
+
Resume from durable state after interruption. Obey `WP_GOAL_DISABLE=1`, iteration/error/validation caps, and repository ownership. The pipeline must never self-approve, directly mark tasks done, bypass evidence, or push to main. Keep every loop bounded and never increase timeouts.
|
|
@@ -0,0 +1,27 @@
|
|
|
1
|
+
---
|
|
2
|
+
type: skill
|
|
3
|
+
slug: goal
|
|
4
|
+
title: Goal
|
|
5
|
+
status: active
|
|
6
|
+
scope: repo
|
|
7
|
+
applies_to: [agents]
|
|
8
|
+
related: [autopilot, plan-refine, verify]
|
|
9
|
+
created: "2026-07-10"
|
|
10
|
+
last_reviewed: "2026-07-10"
|
|
11
|
+
name: goal
|
|
12
|
+
description: "Run durable multi-blueprint goals through evidence-gated handoffs."
|
|
13
|
+
argument-hint: "<new|status|handoff|run|cancel> [brief]"
|
|
14
|
+
---
|
|
15
|
+
|
|
16
|
+
# Goal
|
|
17
|
+
|
|
18
|
+
Use `wp goal` as the durable controller for work spanning one or more blueprints.
|
|
19
|
+
|
|
20
|
+
## Protocol
|
|
21
|
+
|
|
22
|
+
1. Create or resume the goal in a managed blueprint owner worktree.
|
|
23
|
+
2. Read `wp goal status`, execute only the current phase handoff, then call `wp goal run` again.
|
|
24
|
+
3. Complete blueprint tasks only through `wp_blueprint_task_verify` with passing evidence.
|
|
25
|
+
4. Stop when the controller reports `complete`, `cancelled`, blocked guard exhaustion, or `WP_GOAL_DISABLE=1`.
|
|
26
|
+
|
|
27
|
+
The blueprint engine owns task status and approval gates. The goal controller must never self-approve, directly mark a task done, bypass review provenance, or mutate the primary checkout. Keep every loop bounded and never increase timeouts.
|