@webpresso/codex-plugin 3.3.1 → 3.3.3
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/package.json
CHANGED
|
@@ -2,7 +2,7 @@
|
|
|
2
2
|
"schemaVersion": 1,
|
|
3
3
|
"host": "codex",
|
|
4
4
|
"packageName": "@webpresso/codex-plugin",
|
|
5
|
-
"packageVersion": "3.3.
|
|
5
|
+
"packageVersion": "3.3.3",
|
|
6
6
|
"runtimeDirs": [".codex/skills", ".agents/skills"],
|
|
7
7
|
"skills": {
|
|
8
8
|
"ai-deslop": {
|
|
@@ -87,10 +87,10 @@
|
|
|
87
87
|
"digest": "sha256:bed3684a850651534f315d355cdeae50d951c429817df0f105cccdb7f4e1561b"
|
|
88
88
|
},
|
|
89
89
|
"ultragoal": {
|
|
90
|
-
"digest": "sha256:
|
|
90
|
+
"digest": "sha256:7ed3e728027a9d1047a1dd030f88f23a6120450bf336fa73b1c9037857ab56ee"
|
|
91
91
|
},
|
|
92
92
|
"verify": {
|
|
93
|
-
"digest": "sha256:
|
|
93
|
+
"digest": "sha256:bb22b3c997ef3b003e5558c01f0b1e844158a48f1bb5d7169851e5b0b343cd64"
|
|
94
94
|
}
|
|
95
95
|
}
|
|
96
96
|
}
|
|
@@ -109,24 +109,11 @@ Resolve models from the live catalog every time:
|
|
|
109
109
|
opencode models opencode-go
|
|
110
110
|
```
|
|
111
111
|
|
|
112
|
-
|
|
113
|
-
|
|
114
|
-
|
|
115
|
-
|
|
116
|
-
|
|
117
|
-
| Coding / implementation quality | Kimi **Code** first (`…kimi…-code`) | Coding quality lane |
|
|
118
|
-
| Cheap recheck / high-availability follow-up | DeepSeek **Flash** (`…deepseek…-flash`) | Fast secondary pass |
|
|
119
|
-
| Aggregate / unspecified | Follow `opencode-go` skill order: Kimi Code → DeepSeek Pro → DeepSeek Flash → remaining families | Default outside-voice lane |
|
|
120
|
-
|
|
121
|
-
Resolve the actual model ID with catalog greps (never paste a fixed model string into agent instructions as the only option):
|
|
122
|
-
|
|
123
|
-
```bash
|
|
124
|
-
CATALOG=$(opencode models opencode-go)
|
|
125
|
-
# plan critique example:
|
|
126
|
-
MODEL=$(echo "$CATALOG" | grep '^opencode-go/deepseek' | grep -- '-pro$' | sort -V | tail -1)
|
|
127
|
-
# coding example:
|
|
128
|
-
# MODEL=$(echo "$CATALOG" | grep '^opencode-go/kimi' | grep -- '-code$' | sort -V | tail -1)
|
|
129
|
-
```
|
|
112
|
+
**Selection order and the purpose→family mapping live in one place: the committed
|
|
113
|
+
reviewer policy (`packages/workflow-skills/src/opencode-reviewer-policy.ts`),
|
|
114
|
+
rendered into the `opencode-go` skill and drift-gated against it.** Follow that
|
|
115
|
+
skill; do not restate the ordering here. A second copy of a preference list is a
|
|
116
|
+
second thing to update, and the copy that goes stale is the one an agent reads.
|
|
130
117
|
|
|
131
118
|
Treat a provider-wide usage limit as terminal for every OpenCode Go model in the current review window; do not hop families to evade a usage block on the same provider.
|
|
132
119
|
|
package/skills/verify/SKILL.md
CHANGED
|
@@ -35,18 +35,18 @@ the preferred skill (`claude` / `codex` / `grok` / one OpenCode Go), then stops.
|
|
|
35
35
|
|
|
36
36
|
## Tooling map
|
|
37
37
|
|
|
38
|
-
| Gate | Prefer
|
|
39
|
-
| ------------------------- |
|
|
40
|
-
| Lint / types / tests / QA | `wp_lint`, `wp_typecheck`, `wp_test`, `wp_qa`
|
|
41
|
-
| Format | `wp_format`
|
|
42
|
-
| Audits | `wp_audit`, `wp_audits`
|
|
43
|
-
| Blueprint | blueprint MCP tools + `wp audit blueprint-lifecycle`
|
|
44
|
-
| PR / CI wait | `wp_pr_status`, `wp_pr_wait` (not multi-sleep loops)
|
|
45
|
-
| Release | `wp_release_readiness`
|
|
46
|
-
| Large output | `wp_session_execute` / `batch_execute` / `execute_file`
|
|
47
|
-
| Continuity | `wp_session_capture` for verify summary; avoid paste-back walls
|
|
48
|
-
| E2E / act | `wp_e2e`, `wp_ci_act` when the change requires them
|
|
49
|
-
| Outside authority | `wp review gate` (not raw multi-agent review swarms)
|
|
38
|
+
| Gate | Prefer |
|
|
39
|
+
| ------------------------- | --------------------------------------------------------------------------------------- |
|
|
40
|
+
| Lint / types / tests / QA | `wp_lint`, `wp_typecheck`, `wp_test`, `wp_qa` |
|
|
41
|
+
| Format | `wp_format` |
|
|
42
|
+
| Audits | `wp_audit`, `wp_audits` |
|
|
43
|
+
| Blueprint | blueprint MCP tools + `wp audit blueprint-lifecycle` |
|
|
44
|
+
| PR / CI wait | `wp_pr_status`, `wp_pr_wait` (not multi-sleep loops) |
|
|
45
|
+
| Release | `wp_release_progress` for live Release phases; `wp_release_readiness` for package gates |
|
|
46
|
+
| Large output | `wp_session_execute` / `batch_execute` / `execute_file` |
|
|
47
|
+
| Continuity | `wp_session_capture` for verify summary; avoid paste-back walls |
|
|
48
|
+
| E2E / act | `wp_e2e`, `wp_ci_act` when the change requires them |
|
|
49
|
+
| Outside authority | `wp review gate` (not raw multi-agent review swarms) |
|
|
50
50
|
|
|
51
51
|
## Iron law — evidence before claims
|
|
52
52
|
|