@nolrm/contextkit 0.12.19 → 0.12.20
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
|
@@ -153,6 +153,7 @@ ContextKit installs reusable slash commands for supported platforms:
|
|
|
153
153
|
| `/squad-dev` | Implement code following the architect plan |
|
|
154
154
|
| `/squad-test` | Write and run tests against acceptance criteria |
|
|
155
155
|
| `/squad-review` | Review the full pipeline and give a verdict |
|
|
156
|
+
| `/squad-doc` | Create companion `.md` files for new/modified code after review passes |
|
|
156
157
|
| `/squad-auto` | Auto-run the full pipeline after kickoff (recommended, sequential) |
|
|
157
158
|
| `/squad-auto-parallel` | Auto-run the pipeline in parallel using Claude Code agents (Claude Code only) |
|
|
158
159
|
| `/ck` | Health check — verify setup, standards, and integrations |
|
|
@@ -177,18 +178,20 @@ The squad workflow turns a single AI session into a structured multi-role pipeli
|
|
|
177
178
|
| 3 | Developer | `/squad-dev` | Implements the code following the architect's plan |
|
|
178
179
|
| 4 | Tester | `/squad-test` | Writes and runs tests against the PO's acceptance criteria |
|
|
179
180
|
| 5 | Reviewer | `/squad-review` | Reviews everything and gives a PASS or NEEDS-WORK verdict |
|
|
181
|
+
| 6 | Doc Writer | `/squad-doc` | Creates companion `.md` files for every new/modified code file |
|
|
180
182
|
|
|
181
183
|
### Single-Task Flow
|
|
182
184
|
|
|
183
185
|
```bash
|
|
184
186
|
/squad "add dark mode support" # PO writes the spec
|
|
185
187
|
|
|
186
|
-
/squad-auto # Auto-runs architect → dev → test → review (recommended)
|
|
188
|
+
/squad-auto # Auto-runs architect → dev → test → review → doc (recommended)
|
|
187
189
|
# — or step through manually —
|
|
188
190
|
/squad-architect # Architect designs the plan
|
|
189
191
|
/squad-dev # Dev implements the code
|
|
190
192
|
/squad-test # Tester writes and runs tests
|
|
191
193
|
/squad-review # Reviewer gives the verdict
|
|
194
|
+
/squad-doc # Doc Writer creates companion .md files
|
|
192
195
|
```
|
|
193
196
|
|
|
194
197
|
### Batch Flow
|
|
@@ -200,7 +203,7 @@ Pass multiple tasks to `/squad` and it automatically runs in batch mode:
|
|
|
200
203
|
# PO writes specs for all three tasks
|
|
201
204
|
|
|
202
205
|
/squad-auto
|
|
203
|
-
# Runs Architect → Dev → Test → Review for each task sequentially
|
|
206
|
+
# Runs Architect → Dev → Test → Review → Doc for each task sequentially
|
|
204
207
|
```
|
|
205
208
|
|
|
206
209
|
**Parallel mode (Claude Code only):** Use `/squad-auto-parallel` instead of `/squad-auto` to spawn parallel subagents — one per task per phase — so all tasks progress simultaneously rather than one at a time.
|
|
@@ -210,7 +213,7 @@ Pass multiple tasks to `/squad` and it automatically runs in batch mode:
|
|
|
210
213
|
|
|
211
214
|
/squad-auto-parallel
|
|
212
215
|
# Phase 1: architect agents for all 3 tasks run in parallel
|
|
213
|
-
# Phase 2: dev→test→review pipeline runs in parallel per task
|
|
216
|
+
# Phase 2: dev→test→review pipeline runs in parallel per task; Phase 3: doc runs sequentially
|
|
214
217
|
```
|
|
215
218
|
|
|
216
219
|
**Model routing (Claude Code only):** Set `model_routing: true` in `.contextkit/squad/config.md` to have `/squad-auto` automatically use Claude Haiku for Dev and Test phases. Architect and Review always run on your primary model. Saves ~35% tokens with no quality loss — the standards files and Review gate maintain quality.
|
|
@@ -331,7 +334,7 @@ ck note "AI issue" --category "AI Behavior" --priority HIGH
|
|
|
331
334
|
|
|
332
335
|
# Squad — multi-role AI pipeline (slash commands in your AI tool)
|
|
333
336
|
/squad "add dark mode" # PO writes spec
|
|
334
|
-
/squad-auto # runs architect → dev → test → review hands-free
|
|
337
|
+
/squad-auto # runs architect → dev → test → review → doc hands-free
|
|
335
338
|
/squad-reset # clear stuck or mixed squad state
|
|
336
339
|
|
|
337
340
|
# Observability
|
package/lib/commands/install.js
CHANGED
|
@@ -734,6 +734,10 @@ Any design decisions, trade-offs, or open questions to resolve before coding.
|
|
|
734
734
|
`${this.repoUrl}/commands/squad-reset.md`,
|
|
735
735
|
'.contextkit/commands/squad-reset.md'
|
|
736
736
|
);
|
|
737
|
+
await this.downloadManager.downloadFile(
|
|
738
|
+
`${this.repoUrl}/commands/squad-doc.md`,
|
|
739
|
+
'.contextkit/commands/squad-doc.md'
|
|
740
|
+
);
|
|
737
741
|
await this.downloadManager.downloadFile(
|
|
738
742
|
`${this.repoUrl}/commands/health-check.md`,
|
|
739
743
|
'.contextkit/commands/health-check.md'
|
package/lib/commands/update.js
CHANGED
|
@@ -291,6 +291,10 @@ class UpdateCommand {
|
|
|
291
291
|
`${this.repoUrl}/commands/squad-reset.md`,
|
|
292
292
|
'.contextkit/commands/squad-reset.md'
|
|
293
293
|
);
|
|
294
|
+
await this.downloadManager.downloadFile(
|
|
295
|
+
`${this.repoUrl}/commands/squad-doc.md`,
|
|
296
|
+
'.contextkit/commands/squad-doc.md'
|
|
297
|
+
);
|
|
294
298
|
await this.downloadManager.downloadFile(
|
|
295
299
|
`${this.repoUrl}/commands/health-check.md`,
|
|
296
300
|
'.contextkit/commands/health-check.md'
|
|
@@ -25,6 +25,7 @@ class ClaudeIntegration extends BaseIntegration {
|
|
|
25
25
|
'.claude/commands/squad-auto.md',
|
|
26
26
|
'.claude/commands/squad-auto-parallel.md',
|
|
27
27
|
'.claude/commands/squad-reset.md',
|
|
28
|
+
'.claude/commands/squad-doc.md',
|
|
28
29
|
'.claude/commands/spec.md',
|
|
29
30
|
'.claude/commands/ck.md',
|
|
30
31
|
];
|
|
@@ -216,7 +217,7 @@ Read \`.contextkit/commands/squad.md\` and execute the squad kickoff workflow.
|
|
|
216
217
|
|
|
217
218
|
Create the handoff file and write the PO spec for the given task. Pass the user's task description as the input.
|
|
218
219
|
|
|
219
|
-
After kickoff, run \`/squad-auto\` to auto-run the full pipeline hands-free, or step through manually with \`/squad-architect\` → \`/squad-dev\` → \`/squad-test\` → \`/squad-review\`.
|
|
220
|
+
After kickoff, run \`/squad-auto\` to auto-run the full pipeline hands-free, or step through manually with \`/squad-architect\` → \`/squad-dev\` → \`/squad-test\` → \`/squad-review\` → \`/squad-doc\`.
|
|
220
221
|
`);
|
|
221
222
|
|
|
222
223
|
await this.writeGeneratedFile('.claude/commands/squad-architect.md', `# Squad — Architect (manual step 1/4)
|
|
@@ -266,6 +267,13 @@ Spawn one subagent per task per phase using the Task tool, so all tasks progress
|
|
|
266
267
|
Read \`.contextkit/commands/squad-reset.md\` and execute the reset workflow.
|
|
267
268
|
|
|
268
269
|
Delete the current squad state (.contextkit/squad/) so you can start fresh. Use when the squad folder is in a mixed or stuck state.
|
|
270
|
+
`);
|
|
271
|
+
|
|
272
|
+
await this.writeGeneratedFile('.claude/commands/squad-doc.md', `# Squad — Doc (manual step 5/5)
|
|
273
|
+
|
|
274
|
+
Read \`.contextkit/commands/squad-doc.md\` and execute the doc workflow.
|
|
275
|
+
|
|
276
|
+
After review passes, create or update companion .md files for every new/modified code file in this task. Use \`/squad-auto\` instead to run all steps automatically.
|
|
269
277
|
`);
|
|
270
278
|
|
|
271
279
|
await this.writeGeneratedFile('.claude/commands/ck.md', `# ContextKit Health Check
|
|
@@ -298,6 +306,7 @@ Check project setup, standards status, and integrations. Report what needs atten
|
|
|
298
306
|
console.log(chalk.dim(' /squad-review — Review and write verdict'));
|
|
299
307
|
console.log(chalk.dim(' /squad-auto — Auto-run full pipeline (recommended)'));
|
|
300
308
|
console.log(chalk.dim(' /squad-auto-parallel — Auto-run pipeline in parallel (batch, fastest)'));
|
|
309
|
+
console.log(chalk.dim(' /squad-doc — Document changes after review passes'));
|
|
301
310
|
console.log(chalk.dim(''));
|
|
302
311
|
console.log(chalk.dim(' Health check:'));
|
|
303
312
|
console.log(chalk.dim(' /ck — Check project setup and standards status'));
|
|
@@ -0,0 +1,29 @@
|
|
|
1
|
+
# notifier.js
|
|
2
|
+
|
|
3
|
+
Thin wrapper around `update-notifier` for passive npm update notifications in the ContextKit CLI.
|
|
4
|
+
|
|
5
|
+
## Purpose
|
|
6
|
+
|
|
7
|
+
Checks npm once every 24 hours for a newer version of `@nolrm/contextkit`. If an update is available, prints a non-blocking one-liner after the command completes. Suppressed automatically in CI environments (`CI=true`).
|
|
8
|
+
|
|
9
|
+
## Exports
|
|
10
|
+
|
|
11
|
+
### `checkForUpdates()`
|
|
12
|
+
|
|
13
|
+
Initialises `update-notifier` with the package metadata from `package.json` and calls `.notify()`. Errors are silently swallowed — a notifier failure must never crash the CLI.
|
|
14
|
+
|
|
15
|
+
## Usage
|
|
16
|
+
|
|
17
|
+
Called once in `bin/contextkit.js` before `program.parse()`:
|
|
18
|
+
|
|
19
|
+
```javascript
|
|
20
|
+
const { checkForUpdates } = require('../lib/utils/notifier');
|
|
21
|
+
checkForUpdates();
|
|
22
|
+
program.parse(process.argv);
|
|
23
|
+
```
|
|
24
|
+
|
|
25
|
+
## Edge Cases
|
|
26
|
+
|
|
27
|
+
- `update-notifier` caches the check result for 24 hours — only one npm request per day per machine.
|
|
28
|
+
- Suppressed in CI via `update-notifier`'s built-in CI detection (`CI`, `CONTINUOUS_INTEGRATION`, `BUILD_NUMBER` env vars).
|
|
29
|
+
- Uses `update-notifier@5.1.0` — the last CommonJS-compatible release (v6+ is ESM-only).
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@nolrm/contextkit",
|
|
3
|
-
"version": "0.12.
|
|
3
|
+
"version": "0.12.20",
|
|
4
4
|
"description": "ContextKit - Context Engineering for AI Development. Provide rich context to AI through structured MD files with standards, code guides, and documentation. Works with Cursor, Claude, Aider, VS Code Copilot, and more.",
|
|
5
5
|
"main": "lib/index.js",
|
|
6
6
|
"bin": {
|