@lunora/cli 1.0.0-alpha.160 → 1.0.0-alpha.162
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 +11 -11
- package/skills/README.md +9 -5
- package/skills/lunora/SKILL.md +17 -0
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@lunora/cli",
|
|
3
|
-
"version": "1.0.0-alpha.
|
|
3
|
+
"version": "1.0.0-alpha.162",
|
|
4
4
|
"description": "The Lunora CLI: init, dev, deploy, codegen, migrate, seed, doctor, insights, logs, registry, and the rest of the project commands",
|
|
5
5
|
"keywords": [
|
|
6
6
|
"agent-skills",
|
|
@@ -52,16 +52,16 @@
|
|
|
52
52
|
},
|
|
53
53
|
"dependencies": {
|
|
54
54
|
"@bomb.sh/tab": "0.0.21",
|
|
55
|
-
"@lunora/advisor": "1.0.0-alpha.
|
|
56
|
-
"@lunora/bindings": "1.0.0-alpha.
|
|
57
|
-
"@lunora/codegen": "1.0.0-alpha.
|
|
58
|
-
"@lunora/config": "1.0.0-alpha.
|
|
59
|
-
"@lunora/container": "1.0.0-alpha.
|
|
60
|
-
"@lunora/d1": "1.0.0-alpha.
|
|
61
|
-
"@lunora/errors": "1.0.0-alpha.
|
|
62
|
-
"@lunora/mcp": "1.0.0-alpha.
|
|
63
|
-
"@lunora/runtime": "1.0.0-alpha.
|
|
64
|
-
"@lunora/seed": "1.0.0-alpha.
|
|
55
|
+
"@lunora/advisor": "1.0.0-alpha.76",
|
|
56
|
+
"@lunora/bindings": "1.0.0-alpha.28",
|
|
57
|
+
"@lunora/codegen": "1.0.0-alpha.106",
|
|
58
|
+
"@lunora/config": "1.0.0-alpha.135",
|
|
59
|
+
"@lunora/container": "1.0.0-alpha.30",
|
|
60
|
+
"@lunora/d1": "1.0.0-alpha.75",
|
|
61
|
+
"@lunora/errors": "1.0.0-alpha.21",
|
|
62
|
+
"@lunora/mcp": "1.0.0-alpha.68",
|
|
63
|
+
"@lunora/runtime": "1.0.0-alpha.61",
|
|
64
|
+
"@lunora/seed": "1.0.0-alpha.70",
|
|
65
65
|
"@visulima/cerebro": "3.0.0",
|
|
66
66
|
"@visulima/error": "6.0.0",
|
|
67
67
|
"@visulima/fs": "5.0.5",
|
package/skills/README.md
CHANGED
|
@@ -26,18 +26,22 @@ own directory:
|
|
|
26
26
|
| `lunora-deploy` | Deploying to Cloudflare — wrangler bindings, secrets, the gate. |
|
|
27
27
|
| `lunora-performance-audit` | `lunora insights`, scans, indexes, OCC conflicts, sharding/`.global()`. |
|
|
28
28
|
|
|
29
|
-
These are mirrored into `.agents/skills/` and `.claude/skills
|
|
30
|
-
so agents working inside this repo pick them up directly
|
|
31
|
-
|
|
29
|
+
These are mirrored (via symlinks) into `.agents/skills/` and `.claude/skills/`,
|
|
30
|
+
so agents working inside this repo pick them up directly, and into
|
|
31
|
+
`plugins/lunora/skills/`, the payload of the Claude Code / Codex plugin. The
|
|
32
|
+
source of truth lives here so the published `@lunora/cli` tarball carries them —
|
|
33
|
+
`lunora rules install` copies this directory into a consumer project.
|
|
32
34
|
|
|
33
|
-
**Adding a skill?** Create it here, then mirror it through
|
|
35
|
+
**Adding a skill?** Create it here, then mirror it through all three hops:
|
|
34
36
|
|
|
35
37
|
```bash
|
|
36
38
|
ln -s ../../packages/cli/skills/<name> .agents/skills/<name>
|
|
37
39
|
ln -s ../../.agents/skills/<name> .claude/skills/<name>
|
|
40
|
+
ln -s ../../../packages/cli/skills/<name> plugins/lunora/skills/<name>
|
|
38
41
|
```
|
|
39
42
|
|
|
40
43
|
`scripts/check-skill-mirrors.js` (run on every `pnpm install`) fails if a mirror
|
|
41
44
|
is missing, dangling, or has become a real directory — a copied directory keeps
|
|
42
45
|
resolving while it silently drifts from the source, so it is treated as an
|
|
43
|
-
error rather than a warning.
|
|
46
|
+
error rather than a warning. Forget a hop and it prints the exact `ln -s` for
|
|
47
|
+
the one you missed, so `pnpm install` is the shortest path to the right command.
|
package/skills/lunora/SKILL.md
CHANGED
|
@@ -63,6 +63,23 @@ After codegen is green, use the most specific Lunora skill for the task:
|
|
|
63
63
|
- Investigating performance, scan, or write-conflict issues:
|
|
64
64
|
`lunora-performance-audit`
|
|
65
65
|
|
|
66
|
+
**Reviewing or auditing existing `lunora/` code** has no skill of its own,
|
|
67
|
+
because two commands do the work mechanically — run them before reading
|
|
68
|
+
anything:
|
|
69
|
+
|
|
70
|
+
```bash
|
|
71
|
+
lunora verify # wrangler config + codegen dry-run + tsc --noEmit
|
|
72
|
+
lunora advisor # the static + runtime lint set, scored per procedure
|
|
73
|
+
```
|
|
74
|
+
|
|
75
|
+
`lunora advisor` is the security/quality review: RLS coverage, ownership and
|
|
76
|
+
identity checks, fail-open guards, input validators, unindexed reads,
|
|
77
|
+
non-deterministic calls in queries, SQL interpolation, and leaked secrets. It
|
|
78
|
+
prints each finding's own lint id and category, so read its output rather than
|
|
79
|
+
grepping for rule names. Use `--entry <file>#<export>` to inspect one procedure
|
|
80
|
+
and `--min-score` / `--baseline` to gate CI. Only after it is clean is a by-hand
|
|
81
|
+
pass over `lunora-functions` worth the tokens.
|
|
82
|
+
|
|
66
83
|
If one of those clearly matches the user's goal, switch to it instead of staying
|
|
67
84
|
in this skill.
|
|
68
85
|
|