@weppy/ralph 0.1.0
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.
- package/README.md +59 -0
- package/dist/cli.js +2248 -0
- package/dist/index.js +28 -0
- package/package.json +43 -0
- package/specs/00-/352/260/234/354/232/224.md +40 -0
- package/specs/10-/354/225/204/355/202/244/355/205/215/354/262/230.md +89 -0
- package/specs/20-/354/203/201/355/203/234/354/231/200-/354/213/244/355/226/211.md +161 -0
- package/specs/30-/354/235/270/355/204/260/355/216/230/354/235/264/354/212/244.md +137 -0
package/README.md
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
1
|
+
# @weppy/ralph
|
|
2
|
+
|
|
3
|
+
`@weppy/ralph` is an independent orchestrator package for fresh-context AI agent execution.
|
|
4
|
+
|
|
5
|
+
## Scope
|
|
6
|
+
|
|
7
|
+
- File-backed job state under `.ralph-cache/` by default
|
|
8
|
+
- Adapter-based execution for Codex and Claude Code
|
|
9
|
+
- Ralph-owned validation, retry, completion, and resume rules
|
|
10
|
+
- Shared core with both CLI and MCP access layers
|
|
11
|
+
- Input files and images are passed as path references for the agent to inspect directly
|
|
12
|
+
- Run state and placeholder logs are written before agent execution starts
|
|
13
|
+
|
|
14
|
+
## Usage
|
|
15
|
+
|
|
16
|
+
```bash
|
|
17
|
+
ralph start \
|
|
18
|
+
--title "Implement auth flow" \
|
|
19
|
+
--agent custom-command \
|
|
20
|
+
--workspace /path/to/project \
|
|
21
|
+
--input docs/spec.md \
|
|
22
|
+
--validate-cmd "npm test" \
|
|
23
|
+
--max-retries 1
|
|
24
|
+
|
|
25
|
+
ralph status --workspace /path/to/project --json
|
|
26
|
+
ralph resume --workspace /path/to/project --max-iterations 5 --json
|
|
27
|
+
ralph result --workspace /path/to/project
|
|
28
|
+
ralph cancel --workspace /path/to/project
|
|
29
|
+
```
|
|
30
|
+
|
|
31
|
+
## Adapters
|
|
32
|
+
|
|
33
|
+
- `codex`: runs `codex exec`
|
|
34
|
+
- `claude-code`: runs `claude -p`
|
|
35
|
+
- `custom-command`: runs the shell command from `RALPH_CUSTOM_AGENT_COMMAND`
|
|
36
|
+
|
|
37
|
+
`custom-command` receives these environment variables:
|
|
38
|
+
|
|
39
|
+
- `RALPH_WORKSPACE_PATH`
|
|
40
|
+
- `RALPH_PROMPT_PATH`
|
|
41
|
+
- `RALPH_OUTPUT_PATH`
|
|
42
|
+
- `RALPH_RUN_DIRECTORY_PATH`
|
|
43
|
+
|
|
44
|
+
The command must write a valid `result.json` payload to `RALPH_OUTPUT_PATH`.
|
|
45
|
+
`RALPH_OUTPUT_PATH` points directly to the run's `result.json` file.
|
|
46
|
+
|
|
47
|
+
## Specs
|
|
48
|
+
|
|
49
|
+
- [개요](./specs/00-%EA%B0%9C%EC%9A%94.md)
|
|
50
|
+
- [아키텍처](./specs/10-%EC%95%84%ED%82%A4%ED%85%8D%EC%B2%98.md)
|
|
51
|
+
- [상태와 실행](./specs/20-%EC%83%81%ED%83%9C%EC%99%80-%EC%8B%A4%ED%96%89.md)
|
|
52
|
+
- [인터페이스](./specs/30-%EC%9D%B8%ED%84%B0%ED%8E%98%EC%9D%B4%EC%8A%A4.md)
|
|
53
|
+
- [구현 계획](./specs/40-%EA%B5%AC%ED%98%84%EA%B3%84%ED%9A%8D.md)
|
|
54
|
+
|
|
55
|
+
## Scripts
|
|
56
|
+
|
|
57
|
+
- `npm run check`
|
|
58
|
+
- `npm run build`
|
|
59
|
+
- `npm test`
|