cc-codex-run 0.1.1
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/CHANGELOG.md +22 -0
- package/LICENSE +21 -0
- package/README.md +145 -0
- package/README.zh-CN.md +167 -0
- package/package.json +51 -0
- package/plugin/.claude-plugin/plugin.json +17 -0
- package/plugin/skills/codex/SKILL.md +85 -0
- package/plugin/skills/codex/roles/check.toml +11 -0
- package/plugin/skills/codex/roles/critic.toml +8 -0
- package/plugin/skills/codex/roles/explore.toml +9 -0
- package/plugin/skills/codex/roles/implement.toml +12 -0
- package/plugin/skills/codex/schemas/check.json +99 -0
- package/plugin/skills/codex/scripts/THIRD_PARTY_LICENSES.md +158 -0
- package/plugin/skills/codex/scripts/codex-run.mjs +12632 -0
- package/plugin/skills/codex-delegate/SKILL.md +49 -0
|
@@ -0,0 +1,49 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: codex-delegate
|
|
3
|
+
description: Delegation policy for handing work to Codex — implementation of modules whose interface contract is fixed, independent verification, and pre-commit review. Use when deciding whether and how to hand a piece of work to Codex.
|
|
4
|
+
---
|
|
5
|
+
|
|
6
|
+
# Handing work to Codex
|
|
7
|
+
|
|
8
|
+
All three cases dispatch through the `codex` skill.
|
|
9
|
+
|
|
10
|
+
## Scope
|
|
11
|
+
|
|
12
|
+
By default Codex takes:
|
|
13
|
+
|
|
14
|
+
- **Implementation of self-contained modules whose interface contract is already fixed**, typically backend modules (routes, services, data access, CLIs). Work whose correctness depends on look and feel, such as UI, stays with you.
|
|
15
|
+
- **Independent verification** and **pre-commit review**.
|
|
16
|
+
|
|
17
|
+
The user's or project's instructions override this scope. When they widen it (for example, "Codex may do frontend work") or narrow it, follow them.
|
|
18
|
+
|
|
19
|
+
## Implementation
|
|
20
|
+
|
|
21
|
+
1. Before dispatching, write these into the task file:
|
|
22
|
+
- the interface contract (routes, signatures, data shapes, error semantics)
|
|
23
|
+
- file ownership
|
|
24
|
+
- the real entry point that exercises the module
|
|
25
|
+
- shared resources: ports, databases, whether dependencies may be installed
|
|
26
|
+
|
|
27
|
+
If you cannot fill these in, do the work yourself or ask the user.
|
|
28
|
+
2. Use the `implement` role. Dispatch independent modules in parallel in one message, and keep working on other independent things.
|
|
29
|
+
3. When results come back, read the diff, trigger the real entry point once yourself, then integrate.
|
|
30
|
+
|
|
31
|
+
## Independent verification
|
|
32
|
+
|
|
33
|
+
Use it when the user asks, or when the change is high-risk: schema or data migrations, authentication or permission checks, money or billing calculations, concurrency/locking/idempotency, or incompatible changes to an external API. Ordinary changes do not need it.
|
|
34
|
+
|
|
35
|
+
1. Write into the task file:
|
|
36
|
+
- the working directory
|
|
37
|
+
- the scope of the change (files or diff)
|
|
38
|
+
- the expected behavior
|
|
39
|
+
- the real entry point
|
|
40
|
+
|
|
41
|
+
Fill in anything missing before dispatching.
|
|
42
|
+
2. Use the `check` role with `--schema <codex skill dir>/schemas/check.json` (the schema the `codex` skill ships), so the result comes back as structured counterexamples. If Codex wrote the change, verify with a different model than the author's (`-m`).
|
|
43
|
+
3. After fixing counterexamples, resume the same thread by id so it re-verifies them.
|
|
44
|
+
|
|
45
|
+
## Pre-commit review
|
|
46
|
+
|
|
47
|
+
1. Run it when the user asks, or before committing or merging a batch of non-trivial changes. Use the `codex` skill's Review section: the default for uncommitted changes, `--base <branch>` for a branch. Put known risk areas in the focus file.
|
|
48
|
+
2. Review is read-only and runs no code. To find counterexamples by actually exercising the change, use independent verification instead.
|
|
49
|
+
3. Go through each finding: fix the valid ones, and explain why the others do not hold.
|