oh-my-workflow 0.2.0 → 0.2.1
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 +2 -2
- package/package.json +1 -1
- package/skill/SKILL.md +26 -13
package/README.md
CHANGED
|
@@ -118,8 +118,8 @@ A node is a coding agent driven through its headless prompt→result CLI.
|
|
|
118
118
|
| adapter | status | notes |
|
|
119
119
|
|---|---|---|
|
|
120
120
|
| **fake** | built-in, free, deterministic | the no-key demo engine and test double |
|
|
121
|
-
| **claude** | **full** (live-verified, 2.1.
|
|
122
|
-
| **codex** | **experimental** (live-verified, 0.137.
|
|
121
|
+
| **claude** | **full** (live-verified, 2.1.x) | `claude -p --output-format json`; `--resume` powers in-session schema self-repair |
|
|
122
|
+
| **codex** | **experimental** (live-verified, 0.137.x) | `codex exec --json`; **no cost field**; tolerates malformed JSONL ([openai/codex#15451](https://github.com/openai/codex/issues/15451)) and fails *actionably* |
|
|
123
123
|
| **pi** | planned | not wired yet (`--agent pi` → exit 3 + install hint) |
|
|
124
124
|
| **kiro** | not a fit | its CLI is an IDE launcher (open files/diffs), no headless prompt→result interface |
|
|
125
125
|
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "oh-my-workflow",
|
|
3
|
-
"version": "0.2.
|
|
3
|
+
"version": "0.2.1",
|
|
4
4
|
"description": "Run coding-agent CLIs (claude -p / codex exec) as nodes in a plain-JS workflow. The thin deterministic glue.",
|
|
5
5
|
"type": "module",
|
|
6
6
|
"author": "Dongwook Kim (https://github.com/domuk-k)",
|
package/skill/SKILL.md
CHANGED
|
@@ -34,22 +34,35 @@ bad node never crashes the run.
|
|
|
34
34
|
single raw LLM API call (that's LangGraph/Mastra territory; an omw node is a
|
|
35
35
|
*whole coding agent*).
|
|
36
36
|
|
|
37
|
-
## The 30-second free demo (no API key)
|
|
37
|
+
## The 30-second free demo (no API key, nothing to clone)
|
|
38
|
+
|
|
39
|
+
omw is on npm, so you can run the whole thing in one line — no install step, no
|
|
40
|
+
key, no cost:
|
|
41
|
+
|
|
42
|
+
```sh
|
|
43
|
+
bunx oh-my-workflow@latest run examples/deep-research --agent fake
|
|
44
|
+
# → {"confirmed":[…],"summary":{…}} exit 0 · no key · no cost · deterministic
|
|
45
|
+
```
|
|
46
|
+
|
|
47
|
+
> Tip: use `@latest`. A bare `bunx oh-my-workflow` can serve a stale cached copy.
|
|
48
|
+
|
|
49
|
+
That single command runs the **whole spine** for you — a fan-out search, a
|
|
50
|
+
pipeline, a scripted schema-fail→self-repair, and a scripted timeout→drop — and
|
|
51
|
+
prints one result JSON. Want to watch it happen? Add `--pretty` for the
|
|
52
|
+
phase/fan-out tree on stderr:
|
|
38
53
|
|
|
39
54
|
```sh
|
|
40
|
-
|
|
41
|
-
bun install
|
|
42
|
-
bun src/cli/omw.ts run examples/deep-research --agent fake
|
|
43
|
-
# → {"confirmed":[…],"summary":{…}} exit 0 · no key · no cost · `--agent fake` is deterministic
|
|
55
|
+
bunx oh-my-workflow@latest run examples/deep-research --agent fake --pretty
|
|
44
56
|
```
|
|
45
57
|
|
|
46
|
-
`--agent fake` is a built-in deterministic adapter
|
|
47
|
-
|
|
48
|
-
|
|
49
|
-
tree on stderr. Swap `--agent claude` once you've run `claude login`.
|
|
58
|
+
`--agent fake` is a built-in, deterministic adapter — it's the no-key demo engine
|
|
59
|
+
and the test double. When you're ready for real work, run `claude login` once and
|
|
60
|
+
swap `--agent fake` → `--agent claude`. Same script, real nodes.
|
|
50
61
|
|
|
51
|
-
>
|
|
52
|
-
>
|
|
62
|
+
> **Reading this as a skill?** You already have it. To install/update it for a
|
|
63
|
+
> coding agent: `bunx oh-my-workflow@latest skill install` (→ `~/.claude/skills/`,
|
|
64
|
+
> or `--project` for one repo); `omw skill path` prints the bundled copy for other
|
|
65
|
+
> hosts. Re-run `skill install` anytime to refresh.
|
|
53
66
|
|
|
54
67
|
---
|
|
55
68
|
|
|
@@ -389,8 +402,8 @@ agents that expose such a CLI can be nodes.
|
|
|
389
402
|
| adapter | status | invoke | structured out | in-session follow-up |
|
|
390
403
|
|---|---|---|---|---|
|
|
391
404
|
| **fake** | built-in, free, deterministic | in-process fixtures | as scripted | yes (fixture) |
|
|
392
|
-
| **claude** | **full** (live-verified, claude 2.1.
|
|
393
|
-
| **codex** | **experimental** (live-verified, codex 0.137.
|
|
405
|
+
| **claude** | **full** (live-verified, claude 2.1.x) | `claude -p <p> --output-format json` | parse `.result` | `--resume` |
|
|
406
|
+
| **codex** | **experimental** (live-verified, codex 0.137.x) | `codex exec --json -s workspace-write` | last `agent_message` from JSONL | `exec resume` |
|
|
394
407
|
| **pi** | planned | `pi --print` | stdout | — |
|
|
395
408
|
| **kiro** | **not a fit** | — | — | — |
|
|
396
409
|
|