AGENTS.md

1<!-- Parent: ../AGENTS.md -->
2<!-- Generated: 2026-01-28 | Updated: 2026-03-02 -->
3 
4# skills
5 
630 skill directories for workflow automation and specialized behaviors.
7 
8## Purpose
9 
10Skills are reusable workflow templates that can be invoked via `/oh-my-claudecode:skill-name`. Each skill provides:
11- Structured prompts for specific workflows
12- Activation triggers (manual or automatic)
13- Integration with execution modes
14 
15## Key Files
16 
17### Execution Mode Skills
18 
19| File | Skill | Purpose |
20|-----------|-------|---------|
21| `autopilot/SKILL.md` | autopilot | Full autonomous execution from idea to working code |
22| `ultrawork/SKILL.md` | ultrawork | Maximum parallel agent execution |
23| `ralph/SKILL.md` | ralph | Persistence until verified complete |
24| `team/SKILL.md` | team | N coordinated agents with task claiming |
25| `ultraqa/SKILL.md` | ultraqa | QA cycling until goal met |
26 
27### Planning Skills
28 
29| File | Skill | Purpose |
30|-----------|-------|---------|
31| `plan/SKILL.md` | omc-plan | Strategic planning with interview workflow |
32| `ralplan/SKILL.md` | ralplan | Iterative planning (Planner+Architect+Critic) with RALPLAN-DR structured deliberation (`--deliberate` for high-risk) |
33| `deep-interview/SKILL.md` | deep-interview | Socratic deep interview with mathematical ambiguity gating (Ouroboros-inspired) |
34| `ralph-init/SKILL.md` | ralph-init | Initialize PRD for structured ralph |
35 
36### Exploration Skills
37 
38| File | Skill | Purpose |
39|-----------|-------|---------|
40| `deepinit/SKILL.md` | deepinit | Generate hierarchical AGENTS.md |
41| `sciomc/SKILL.md` | sciomc | Parallel scientist orchestration |
42 
43### Visual Skills
44 
45| File | Skill | Purpose |
46|-----------|-------|---------|
47| `visual-verdict/SKILL.md` | visual-verdict | Structured visual QA verdict for screenshot/reference comparisons |
48 
49### Utility Skills
50 
51| File | Skill | Purpose |
52|-----------|-------|---------|
53| `ai-slop-cleaner/SKILL.md` | ai-slop-cleaner | Regression-safe cleanup workflow for AI-generated code slop |
54| `skillify/SKILL.md` | skillify | Extract reusable skill from session |
55| `learner/SKILL.md` | learner | Deprecated compatibility alias/internal implementation history for skillify |
56| `ask/SKILL.md` | ask | Ask Claude, Codex, or Gemini via `omc ask` and capture an artifact |
57| `note/SKILL.md` | note | Save notes for compaction resilience |
58| `cancel/SKILL.md` | cancel | Cancel any active OMC mode |
59| `hud/SKILL.md` | hud | Configure HUD display |
60| `omc-doctor/SKILL.md` | omc-doctor | Diagnose installation issues |
61| `setup/SKILL.md` | setup | Unified setup entrypoint for install, diagnostics, and MCP configuration |
62| `omc-setup/SKILL.md` | omc-setup | One-time setup wizard |
63| `omc-help/SKILL.md` | omc-help | Usage guide |
64| `mcp-setup/SKILL.md` | mcp-setup | Configure MCP servers |
65| `skill/SKILL.md` | skill | Manage local skills |
66 
67### Domain Skills
68 
69| File | Skill | Purpose |
70|-----------|-------|---------|
71| `project-session-manager/SKILL.md` | project-session-manager (+ `psm` alias) | Isolated dev environments |
72| `writer-memory/SKILL.md` | writer-memory | Agentic memory for writers |
73| `release/SKILL.md` | release | Generic release assistant — analyzes repo CI/rules, caches in `.omc/RELEASE_RULE.md`, guides the release |
74 
75## For AI Agents
76 
77### Working In This Directory
78 
79#### Skill Template Format
80 
81```markdown
82---
83name: skill-name
84description: Brief description
85triggers:
86  - "keyword1"
87  - "keyword2"
88agent: executor  # Optional: which agent to use
89model: sonnet    # Optional: model override
90pipeline: [skill-name, follow-up-skill]  # Optional: standardized multi-skill flow
91next-skill: follow-up-skill              # Optional: explicit handoff target
92next-skill-args: --direct                # Optional: arguments for the next skill
93handoff: .omc/plans/example.md           # Optional: artifact/context handed to next skill
94---
95 
96# Skill Name
97 
98## Purpose
99What this skill accomplishes.
100 
101## Workflow
1021. Step one
1032. Step two
1043. Step three
105 
106## Usage
107How to invoke this skill.
108 
109## Configuration
110Any configurable options.
111```
112 
113#### Skill Invocation
114 
115```bash
116# Manual invocation
117/oh-my-claudecode:skill-name
118 
119# With arguments
120/oh-my-claudecode:skill-name arg1 arg2
121 
122# Auto-detected from keywords
123&quot;autopilot build me a REST API&quot;  # Triggers autopilot skill
124```
125 
126#### Creating a New Skill
127 
1281. Create `new-skill/SKILL.md` directory and file with YAML frontmatter
1292. Define purpose, workflow, and usage
1303. Add to skill registry (auto-detected from frontmatter)
1314. Optionally add activation triggers
1325. Create corresponding plugin-scoped skill/slash surface via `skills/new-skill/SKILL.md` (and generated artifacts when the build requires them)
1336. Update `docs/REFERENCE.md` (Skills section, count)
1347. If execution mode skill, also create `src/hooks/new-skill/` hook
135 
136### Common Patterns
137 
138**Skill chaining:**
139```markdown
140## Workflow
1411. Invoke `explore` agent for context
1422. Invoke `architect` for analysis
1433. Invoke `executor` for implementation
1444. Invoke `qa-tester` for verification
145```
146 
147If `pipeline` / `next-skill` metadata is present, OMC appends a standardized **Skill Pipeline** handoff block to the rendered skill prompt so downstream steps are explicit.
148 
149**Conditional behavior:**
150```markdown
151## Workflow
1521. Check if tests exist
153   - If yes: Run tests first
154   - If no: Create test plan
1552. Proceed with implementation
156```
157 
158### Testing Requirements
159 
160- Skills are verified via integration tests
161- Test skill invocation with `/oh-my-claudecode:skill-name`
162- Verify trigger keywords activate correct skill
163- For git-related skills, follow `templates/rules/git-workflow.md`
164 
165## Dependencies
166 
167### Internal
168- Loaded by skill bridge (`scripts/build-skill-bridge.mjs`)
169- References agents from `agents/`
170- Uses hooks from `src/hooks/`
171 
172### External
173None - pure markdown files.
174 
175## Skill Categories
176 
177| Category | Skills | Trigger Keywords |
178|----------|--------|------------------|
179| Execution | autopilot, ultrawork, ralph, team, ultraqa | "autopilot", "ulw", "ralph", "team" |
180| Cleanup | ai-slop-cleaner | "deslop", "anti-slop", cleanup/refactor + slop smells |
181| Planning | omc-plan, ralplan, deep-interview, ralph-init | "plan this", "interview me", "ouroboros" |
182| Exploration | deepinit, sciomc, external-context | "deepinit", "research" |
183| Utility | skillify, learner (deprecated alias), note, cancel, hud, setup, omc-doctor, omc-setup, omc-help, mcp-setup | "stop", "cancel" |
184| Domain | psm, writer-memory, release | psm context |
185 
186## Auto-Activation
187 
188Some skills activate automatically based on context:
189 
190| Skill | Auto-Trigger Condition |
191|-------|----------------------|
192| autopilot | "autopilot", "build me", "I want a" |
193| ultrawork | "ulw", "ultrawork" |
194| ralph | "ralph", "don't stop until" |
195| deep-interview | "deep interview", "interview me", "ouroboros", "don't assume" |
196| cancel | "stop", "cancel", "abort" |
197 
198<!-- MANUAL:
199- Team runtime wait semantics: `omc_run_team_wait.timeout_ms` only limits the wait call and does not stop workers.
200- `timeoutSeconds` is removed from `omc_run_team_start`; use explicit `omc_run_team_cleanup` for intentional worker pane termination.
201-->
202