@ouro.bot/cli 0.1.0-alpha.36 → 0.1.0-alpha.38

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.
@@ -1,73 +1,60 @@
1
- # Sub-agents
1
+ # Workflow Helpers
2
2
 
3
- These are source-of-truth workflow definitions (`work-planner`, `work-doer`, `work-merger`) for planning, execution, and merge. They can be consumed either as Claude sub-agents (`.md` files with YAML frontmatter) or as Codex-style skills (`SKILL.md`).
3
+ These files are the source-of-truth workflow helpers for:
4
4
 
5
- ## Installation
5
+ - `work-planner`
6
+ - `work-doer`
7
+ - `work-merger`
6
8
 
7
- ### Claude Code (sub-agents)
9
+ They are written to stay generic enough for different agent shells, while following this repo’s local rules through `AGENTS.md`.
8
10
 
9
- Copy or symlink these files into Claude's sub-agent directory:
11
+ ## What They Do
12
+
13
+ - `work-planner.md`
14
+ Creates and refines planning docs, then converts approved plans into doing docs.
15
+ - `work-doer.md`
16
+ Executes approved doing docs unit by unit with strict validation discipline.
17
+ - `work-merger.md`
18
+ Syncs with `main`, resolves conflicts, creates the PR, handles CI, and merges.
19
+
20
+ ## Important Repo-Specific Truth
21
+
22
+ These helpers do not hardcode task-doc paths. They are expected to read project instructions to discover them.
23
+
24
+ In this repo, that means:
25
+
26
+ - task docs live in `~/AgentBundles/<agent>.ouro/tasks/one-shots/`
27
+ - not inside the repo
28
+
29
+ ## Installing For Claude Code
10
30
 
11
31
  ```bash
12
- # Claude Code
32
+ mkdir -p ~/.claude/agents
13
33
  cp subagents/*.md ~/.claude/agents/
14
- # or
15
- ln -s "$(pwd)"/subagents/*.md ~/.claude/agents/
16
34
  ```
17
35
 
18
- ### Codex / skill-based harnesses
19
-
20
- For tools that support skills but not Claude sub-agents, install these as skills:
36
+ ## Installing For Codex-Style Skills
21
37
 
22
38
  ```bash
23
39
  mkdir -p ~/.codex/skills/work-planner ~/.codex/skills/work-doer ~/.codex/skills/work-merger
24
-
25
- # Hard-link to keep one source of truth
26
- ln -f "$(pwd)/subagents/work-planner.md" ~/.codex/skills/work-planner/SKILL.md
27
- ln -f "$(pwd)/subagents/work-doer.md" ~/.codex/skills/work-doer/SKILL.md
28
- ln -f "$(pwd)/subagents/work-merger.md" ~/.codex/skills/work-merger/SKILL.md
40
+ cp subagents/work-planner.md ~/.codex/skills/work-planner/SKILL.md
41
+ cp subagents/work-doer.md ~/.codex/skills/work-doer/SKILL.md
42
+ cp subagents/work-merger.md ~/.codex/skills/work-merger/SKILL.md
29
43
  ```
30
44
 
31
- **Important:** Hard-links break when editors save by replacing the file (new inode). After editing any `subagents/*.md` file, re-run the `ln -f` command for that file to restore the link. You can verify with `stat -f '%i'` — both files should share the same inode.
32
-
33
- Optional UI metadata:
45
+ If you prefer symlinks or hard-links, that is fine too, but plain copies are easier to reason about and easier to repair when editors replace files.
34
46
 
35
- ```bash
36
- mkdir -p ~/.codex/skills/work-planner/agents ~/.codex/skills/work-doer/agents ~/.codex/skills/work-merger/agents
37
- cat > ~/.codex/skills/work-planner/agents/openai.yaml << 'EOF'
38
- interface:
39
- display_name: "Work Planner"
40
- short_description: "Create and gate planning/doing task docs"
41
- default_prompt: "Use $work-planner to create or update a planning doc, then stop at NEEDS_REVIEW."
42
- EOF
43
- cat > ~/.codex/skills/work-doer/agents/openai.yaml << 'EOF'
44
- interface:
45
- display_name: "Work Doer"
46
- short_description: "Execute approved doing docs with strict TDD"
47
- default_prompt: "Use $work-doer to execute an approved doing doc unit by unit."
48
- EOF
49
- cat > ~/.codex/skills/work-merger/agents/openai.yaml << 'EOF'
50
- interface:
51
- display_name: "Work Merger"
52
- short_description: "Merge feature branch into main via PR after work-doer completes"
53
- default_prompt: "Use $work-merger to merge the current feature branch into main."
54
- EOF
55
- ```
47
+ ## Keeping Local Skill Copies Fresh
56
48
 
57
- Restart the harness after install so new skills are discovered.
49
+ After editing any `subagents/*.md` file, resync your local installed copies.
58
50
 
59
- ## Available sub-agents
51
+ The repo workflow usually checks this with diffs like:
60
52
 
61
- | File | Purpose |
62
- |------|---------|
63
- | `work-planner.md` | Interactive task planner. Generates planning docs through conversation, then converts to doing docs after human approval. |
64
- | `work-doer.md` | Task executor. Reads a doing doc and works through each unit sequentially with strict TDD. |
65
- | `work-merger.md` | Sync-and-merge agent. Merges feature branch into main via PR after work-doer completes. Handles conflicts, CI failures, and race conditions. |
53
+ ```bash
54
+ diff -q ~/.codex/skills/work-planner/SKILL.md subagents/work-planner.md
55
+ diff -q ~/.codex/skills/work-doer/SKILL.md subagents/work-doer.md
56
+ ```
66
57
 
67
- ## Workflow
58
+ ## Restart Behavior
68
59
 
69
- 1. Human describes a task
70
- 2. Agent invokes **work-planner** to create a planning doc → human approves → planner converts to doing doc
71
- 3. Agent invokes **work-doer** to execute the doing doc unit by unit
72
- 4. Each unit is committed independently with progress tracked in the doing doc
73
- 5. Agent invokes **work-merger** to merge the feature branch into main via PR (fetch, merge, resolve conflicts, CI gate, merge PR, cleanup)
60
+ Some tools only discover new skills on startup. If a shell/app does not see updates immediately, restart that shell/app after syncing.