oh-my-claude-sisyphus 1.6.0 → 1.7.0

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.
Files changed (2) hide show
  1. package/README.md +80 -12
  2. package/package.json +1 -1
package/README.md CHANGED
@@ -115,6 +115,7 @@ claude
115
115
  | `/orchestrator <task>` | Complex multi-step task coordination |
116
116
  | `/ralph-loop <task>` | Self-referential loop until task completion |
117
117
  | `/cancel-ralph` | Cancel active Ralph Loop |
118
+ | `/update` | Check for and install updates |
118
119
 
119
120
  ### Examples
120
121
 
@@ -158,6 +159,77 @@ Just include these words anywhere in your prompt:
158
159
 
159
160
  ---
160
161
 
162
+ ## Auto-Update
163
+
164
+ Oh-my-claude-sisyphus includes a silent auto-update system that checks for updates in the background. Updates are applied automatically without interrupting your workflow.
165
+
166
+ To manually check for updates:
167
+ ```bash
168
+ /update
169
+ ```
170
+
171
+ ---
172
+
173
+ ## Hooks System
174
+
175
+ Oh-my-claude-sisyphus includes 18 lifecycle hooks that enhance Claude Code's behavior:
176
+
177
+ ### Core Hooks
178
+
179
+ | Hook | Description |
180
+ |------|-------------|
181
+ | **rules-injector** | Dynamic rules injection with YAML frontmatter parsing |
182
+ | **sisyphus-orchestrator** | Enforces orchestrator behavior and delegation |
183
+ | **auto-slash-command** | Automatic slash command detection and execution |
184
+ | **keyword-detector** | Magic keyword detection (ultrawork, search, analyze) |
185
+ | **ralph-loop** | Self-referential development loop management |
186
+ | **todo-continuation** | Ensures todo list completion |
187
+
188
+ ### Context & Recovery
189
+
190
+ | Hook | Description |
191
+ |------|-------------|
192
+ | **context-window-limit-recovery** | Token limit error handling and recovery |
193
+ | **preemptive-compaction** | Context usage monitoring to prevent limits |
194
+ | **session-recovery** | Session state recovery on crashes |
195
+ | **directory-readme-injector** | README context injection |
196
+
197
+ ### Quality & Validation
198
+
199
+ | Hook | Description |
200
+ |------|-------------|
201
+ | **comment-checker** | BDD detection and directive filtering |
202
+ | **thinking-block-validator** | Extended thinking validation |
203
+ | **empty-message-sanitizer** | Empty message handling |
204
+ | **edit-error-recovery** | Automatic recovery from edit errors |
205
+
206
+ ### Environment & Notifications
207
+
208
+ | Hook | Description |
209
+ |------|-------------|
210
+ | **non-interactive-env** | CI/non-interactive environment handling |
211
+ | **agent-usage-reminder** | Reminder to use specialized agents |
212
+ | **background-notification** | Background task completion notifications |
213
+
214
+ ---
215
+
216
+ ## Builtin Skills
217
+
218
+ Six builtin skills provide specialized capabilities:
219
+
220
+ | Skill | Description |
221
+ |-------|-------------|
222
+ | **sisyphus** | Multi-agent orchestration mode |
223
+ | **orchestrator** | Master coordinator for complex tasks |
224
+ | **ultrawork** | Maximum performance with parallel agents |
225
+ | **ralph-loop** | Self-referential development until completion |
226
+ | **frontend-ui-ux** | Designer-turned-developer UI/UX expertise |
227
+ | **git-master** | Git expert for atomic commits and history |
228
+
229
+ Skills are automatically activated via slash commands or magic keywords.
230
+
231
+ ---
232
+
161
233
  ## The Eleven Agents
162
234
 
163
235
  Claude will automatically delegate to these specialized agents:
@@ -387,29 +459,25 @@ The original oh-my-opencode used multiple AI providers. This port uses Claude mo
387
459
  | Feature | Description |
388
460
  |---------|-------------|
389
461
  | **11 Specialized Agents** | Oracle, Librarian, Explore, Frontend Engineer, Document Writer, Multimodal Looker, Momus, Metis, Orchestrator-Sisyphus, Sisyphus-Junior, Prometheus |
462
+ | **18 Lifecycle Hooks** | rules-injector, sisyphus-orchestrator, auto-slash-command, keyword-detector, ralph-loop, todo-continuation, context-window-limit-recovery, preemptive-compaction, session-recovery, directory-readme-injector, comment-checker, thinking-block-validator, empty-message-sanitizer, edit-error-recovery, non-interactive-env, agent-usage-reminder, background-notification, think-mode |
463
+ | **6 Builtin Skills** | sisyphus, orchestrator, ultrawork, ralph-loop, frontend-ui-ux, git-master |
390
464
  | **Magic Keywords** | `ultrawork`, `search`, `analyze`, `ultrathink` trigger enhanced modes |
391
- | **Slash Commands** | `/sisyphus`, `/sisyphus-default`, `/ultrawork`, `/deepsearch`, `/analyze`, `/plan`, `/review`, `/prometheus`, `/orchestrator`, `/ralph-loop`, `/cancel-ralph` |
465
+ | **Slash Commands** | `/sisyphus`, `/sisyphus-default`, `/ultrawork`, `/deepsearch`, `/analyze`, `/plan`, `/review`, `/prometheus`, `/orchestrator`, `/ralph-loop`, `/cancel-ralph`, `/update` |
466
+ | **Auto-Update System** | Silent background updates with manual `/update` command |
392
467
  | **Configuration System** | JSONC config with multi-source merging |
393
468
  | **Context Injection** | Auto-loads CLAUDE.md and AGENTS.md files |
394
- | **Continuation Enforcement** | System prompt enforces task completion |
469
+ | **Continuation Enforcement** | System prompt and hooks enforce task completion |
470
+ | **Session Recovery** | Automatic state recovery on crashes |
471
+ | **Background Task Manager** | Async agent execution with concurrency limits |
472
+ | **Context Window Recovery** | Multi-stage recovery when hitting token limits |
395
473
  | **MCP Server Configs** | Exa, Context7, grep.app server definitions |
396
474
  | **LSP Tools** | Real LSP server integration with 11 tools |
397
475
  | **AST Tools** | ast-grep integration for structural code search/replace |
398
476
 
399
- #### Partially Implemented ⚠️
400
-
401
- | Feature | What Works | What's Missing |
402
- |---------|------------|----------------|
403
- | **Continuation Hook** | System prompt enforcement | Actual todo state checking |
404
-
405
477
  #### Not Implemented ❌
406
478
 
407
479
  | Feature | Original Capability | Why Not Ported |
408
480
  |---------|---------------------|----------------|
409
- | **22 Lifecycle Hooks** | PreToolUse, PostToolUse, Stop, etc. | Claude Code handles hooks differently |
410
- | **Background Task Manager** | Async agent execution with concurrency limits | Claude Code's Task tool handles this |
411
- | **Context Window Compaction** | Multi-stage recovery when hitting token limits | Claude Code manages this internally |
412
- | **Thinking Block Validator** | Validates AI thinking format | Not needed for Claude |
413
481
  | **Multi-Model Routing** | Route to GPT/Gemini/Grok based on task | Claude-only by design |
414
482
  | **Per-Model Concurrency** | Fine-grained concurrency per provider | Single provider simplifies this |
415
483
  | **Interactive Bash + Tmux** | Advanced terminal with Tmux integration | Standard Bash tool sufficient |
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "oh-my-claude-sisyphus",
3
- "version": "1.6.0",
3
+ "version": "1.7.0",
4
4
  "description": "Multi-agent orchestration system for Claude Agent SDK - Port of oh-my-opencode",
5
5
  "type": "module",
6
6
  "main": "dist/index.js",