ai-sprint-kit 2.1.30 → 2.1.32

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.
Files changed (2) hide show
  1. package/README.md +91 -0
  2. package/package.json +1 -1
package/README.md CHANGED
@@ -105,6 +105,97 @@ Visit [ai-sprint-kit.app.data-espresso.com](https://ai-sprint-kit.app.data-espre
105
105
  /ai-sprint-ralph - Persistent loop until task complete
106
106
  ```
107
107
 
108
+ ## Ralph Loop - Autonomous Task Completion
109
+
110
+ Ralph runs tasks in a persistent loop until completion or max iterations reached. Ideal for migrations, batch fixes, and tasks with clear success criteria.
111
+
112
+ ### Simple Mode (Single Task)
113
+
114
+ ```bash
115
+ # Fix all TypeScript errors
116
+ /ai-sprint-ralph "fix all TypeScript errors" --promise "tsc: 0 errors"
117
+
118
+ # Migrate tests to vitest
119
+ /ai-sprint-ralph "migrate jest to vitest" --max-iterations 30
120
+
121
+ # Add dark mode
122
+ /ai-sprint-ralph "add dark mode toggle" --promise "DARK MODE DONE" --max-iterations 10
123
+ ```
124
+
125
+ ### PRD Mode (Multi-Feature)
126
+
127
+ Create `ai_context/ralph/prd.md`:
128
+
129
+ ```markdown
130
+ # Project: User Authentication
131
+
132
+ ## Features
133
+
134
+ ### 1. Login Form
135
+ - [ ] Add login form component
136
+ - [ ] Implement JWT validation
137
+ - [ ] Add logout functionality
138
+
139
+ ### 2. API Endpoints
140
+ - [ ] Create /api/users endpoint
141
+ - [ ] Add authentication middleware
142
+ ```
143
+
144
+ Then run:
145
+
146
+ ```bash
147
+ /ai-sprint-ralph --prd ai_context/ralph/prd.md --max-iterations 50
148
+ ```
149
+
150
+ Ralph works through each feature, checking off tasks as completed.
151
+
152
+ ### Options
153
+
154
+ | Option | Default | Description |
155
+ |--------|---------|-------------|
156
+ | `--max-iterations N` | 20 | Maximum loop iterations |
157
+ | `--promise STRING` | `<promise>COMPLETED</promise>` | Completion detection string |
158
+ | `--prd FILE` | none | PRD file for multi-feature mode |
159
+ | `--no-auto-commit` | false | Disable git commit after iteration |
160
+ | `--resume` | false | Resume interrupted session |
161
+
162
+ ### Shell Script (Fully Autonomous)
163
+
164
+ For **unattended execution** without approval prompts, use the shell script:
165
+
166
+ ```bash
167
+ # Simple mode
168
+ .claude/scripts/ralph-loop.sh "fix all TypeScript errors" --promise "0 errors"
169
+
170
+ # PRD mode
171
+ .claude/scripts/ralph-loop.sh --prd ai_context/ralph/prd.md --max-iterations 50
172
+
173
+ # Check status
174
+ .claude/scripts/ralph-status.sh
175
+
176
+ # Resume interrupted session
177
+ .claude/scripts/ralph-loop.sh --resume
178
+ ```
179
+
180
+ The shell script uses `--dangerously-skip-permissions` to bypass approval gates - ideal for overnight tasks, migrations, and CI/CD pipelines.
181
+
182
+ ### Writing Good Prompts
183
+
184
+ **Good prompts converge** - each iteration brings you closer to success:
185
+
186
+ ```
187
+ Migrate all test files from Jest to Vitest.
188
+
189
+ SUCCESS CRITERIA:
190
+ 1. All imports use "vitest" not "jest"
191
+ 2. npm run test exits with code 0
192
+ 3. Zero jest dependencies remain
193
+
194
+ When ALL criteria met, output: <promise>MIGRATION COMPLETE</promise>
195
+ ```
196
+
197
+ **Avoid vague goals** like "improve quality" - define measurable completion conditions.
198
+
108
199
  ## Support
109
200
 
110
201
  - GitHub Issues: [ai-sprint-pro/issues](https://github.com/apiasak/ai-sprint-pro/issues)
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "ai-sprint-kit",
3
- "version": "2.1.30",
3
+ "version": "2.1.32",
4
4
  "description": "CLI installer for AI Sprint autonomous development framework - requires license",
5
5
  "main": "lib/installer.js",
6
6
  "bin": {