@khanglvm/relay 0.5.3 → 0.5.4
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 +38 -3
- package/docs/AGENT.md +1 -1
- package/package.json +1 -1
- package/src/cli.js +5 -4
package/README.md
CHANGED
|
@@ -19,14 +19,49 @@ note, Submit — and the agent picks up your answers and keeps going:
|
|
|
19
19
|
## Quick start
|
|
20
20
|
|
|
21
21
|
```sh
|
|
22
|
-
|
|
23
|
-
|
|
24
|
-
|
|
22
|
+
# the rly CLI
|
|
23
|
+
npm i -g @khanglvm/relay
|
|
24
|
+
|
|
25
|
+
# the agent skill → every detected agent (Claude Code, Codex, Cursor, …)
|
|
26
|
+
npx skills add khanglvm/relay --skill relay --all
|
|
27
|
+
|
|
28
|
+
# enforce it (recommended) — a skill alone is an ignorable hint; append relay's
|
|
29
|
+
# short usage rules to your agent's global instructions (per-agent commands below)
|
|
30
|
+
rly skill rules
|
|
25
31
|
```
|
|
26
32
|
|
|
27
33
|
That's it. Next time your agent needs a decision or wants to show you a plan,
|
|
28
34
|
it opens a board like the ones above and waits for your Submit.
|
|
29
35
|
|
|
36
|
+
### Tell your agent to actually use it
|
|
37
|
+
|
|
38
|
+
`rly skill rules` prints relay's short usage rules. A skill is just an ignorable
|
|
39
|
+
hint — these rules in the file your agent *always* reads are what make it reach for
|
|
40
|
+
relay. Append them to your agent's **global** instructions:
|
|
41
|
+
|
|
42
|
+
```sh
|
|
43
|
+
rly skill rules >> ~/.claude/CLAUDE.md # Claude Code
|
|
44
|
+
rly skill rules >> ~/.codex/AGENTS.md # Codex
|
|
45
|
+
rly skill rules >> ~/.gemini/GEMINI.md # Gemini CLI
|
|
46
|
+
rly skill rules >> ~/.config/opencode/AGENTS.md # OpenCode
|
|
47
|
+
```
|
|
48
|
+
|
|
49
|
+
Cursor and GitHub Copilot keep global rules in a settings panel, not a file — run
|
|
50
|
+
`rly skill rules` and paste the output into Cursor's *Settings → Rules → User Rules*
|
|
51
|
+
or Copilot's custom-instructions.
|
|
52
|
+
|
|
53
|
+
<details>
|
|
54
|
+
<summary>More agents</summary>
|
|
55
|
+
|
|
56
|
+
```sh
|
|
57
|
+
rly skill rules >> ~/.codeium/windsurf/memories/global_rules.md # Windsurf
|
|
58
|
+
rly skill rules >> ~/.factory/AGENTS.md # Droid (Factory)
|
|
59
|
+
```
|
|
60
|
+
|
|
61
|
+
Any other agent: run `rly skill rules` and paste the block into whatever file or
|
|
62
|
+
settings panel it reads as global instructions.
|
|
63
|
+
</details>
|
|
64
|
+
|
|
30
65
|
## What it improves
|
|
31
66
|
|
|
32
67
|
| Without relay | With relay |
|
package/docs/AGENT.md
CHANGED
|
@@ -479,4 +479,4 @@ call rather than calling it repeatedly in a loop.
|
|
|
479
479
|
- For sensitive PlantUML diagrams, set `"server": "https://your-server"` to avoid
|
|
480
480
|
sending source to the public plantuml.com server.
|
|
481
481
|
- Bundled universal skill (Claude Code, Codex, any SKILL.md-aware agent):
|
|
482
|
-
`rly skill install` — or `npx skills add khanglvm/relay`.
|
|
482
|
+
`rly skill install` — or `npx skills add khanglvm/relay --skill relay --all`.
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@khanglvm/relay",
|
|
3
|
-
"version": "0.5.
|
|
3
|
+
"version": "0.5.4",
|
|
4
4
|
"description": "Browser-based question boards with rich blocks (markdown, charts, mermaid, tables, code, sandboxed HTML) and element-level annotations for AI coding agents (Claude Code, Codex, …): ask users structured questions, present interactive visuals, collect inline comments, wait for submit, read answers as JSON.",
|
|
5
5
|
"keywords": [
|
|
6
6
|
"ai-agents",
|
package/src/cli.js
CHANGED
|
@@ -782,9 +782,10 @@ function cmdSkill(rest) {
|
|
|
782
782
|
bundled at: ${SKILL_SRC}
|
|
783
783
|
install it: rly skill install # auto-detects ~/.claude, ~/.codex, ~/.agents
|
|
784
784
|
rly skill install --target claude|codex|both|<dir>
|
|
785
|
-
from repo: npx skills add khanglvm/relay
|
|
786
|
-
enforce it: rly skill rules
|
|
787
|
-
#
|
|
785
|
+
from repo: npx skills add khanglvm/relay --skill relay --all
|
|
786
|
+
enforce it: rly skill rules # short always-read rules — paste into the file your
|
|
787
|
+
# agent reads (CLAUDE.md, AGENTS.md, .cursor/rules, …);
|
|
788
|
+
# a skill alone is an ignorable hint
|
|
788
789
|
|
|
789
790
|
The skill teaches your agent the board spec format (questions + rich blocks +
|
|
790
791
|
annotations), the blocking vs --detach patterns, and visualization sizing.
|
|
@@ -853,7 +854,7 @@ NOTES answers & annotations autosave in real time (drafts survive timeout
|
|
|
853
854
|
|
|
854
855
|
AI AGENTS run \`rly agent\` for the complete machine-oriented guide.
|
|
855
856
|
a universal skill is bundled — install with \`rly skill install\`
|
|
856
|
-
(or from the repo: npx skills add khanglvm/relay)`);
|
|
857
|
+
(or from the repo: npx skills add khanglvm/relay --skill relay --all)`);
|
|
857
858
|
return 0;
|
|
858
859
|
}
|
|
859
860
|
|