litclaude-ai 0.3.39 → 0.3.40
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 +9 -1
- package/README.md +6 -6
- package/README_ko-KR.md +6 -6
- package/RELEASE_CHECKLIST.md +7 -7
- package/bin/litclaude-ai.js +87 -1
- package/package.json +1 -1
- package/plugins/litclaude/.claude-plugin/plugin.json +1 -1
- package/plugins/litclaude/bin/litclaude-mcp.js +1 -1
- package/plugins/litclaude/output-styles/asd-ste100.md +46 -0
- package/plugins/litclaude/output-styles/asd-ste100_ko-KR.md +47 -0
- package/plugins/litclaude/output-styles/eli5.md +17 -0
- package/plugins/litclaude/output-styles/eli5_ko-KR.md +19 -0
package/CHANGELOG.md
CHANGED
|
@@ -17,7 +17,15 @@
|
|
|
17
17
|
tools, so the harness read-before-write guard never sees them — copying an
|
|
18
18
|
example literally could clobber a file in the user's working directory.
|
|
19
19
|
|
|
20
|
-
## Unreleased — `lit-comprehend` explainer artifacts
|
|
20
|
+
## Unreleased — output styles enrollment and `lit-comprehend` explainer artifacts
|
|
21
|
+
|
|
22
|
+
- Add output styles enrollment to the installer: a `chooseOutputStyle` prompt
|
|
23
|
+
(mirroring `chooseHudAccent`) lets users pick `off`, `asd-ste100`,
|
|
24
|
+
`asd-ste100-ko`, `eli5`, or `eli5-ko` on every install run, defaulted from
|
|
25
|
+
the prior choice. The installer writes `settings.outputStyle` clobber-safely —
|
|
26
|
+
skipping the write and printing a note when the field was changed outside
|
|
27
|
+
LitClaude — and tracks `outputStyleChoice`/`outputStyleWrittenName` under
|
|
28
|
+
the `litclaude` settings key for idempotent reinstall defaulting.
|
|
21
29
|
|
|
22
30
|
- Add the `lit-comprehend` skill, the `/litclaude:lit-comprehend` command, and the
|
|
23
31
|
`lit-comprehend` / `comprehend` / `lit comprehend` / `$lit-comprehend` /
|
package/README.md
CHANGED
|
@@ -9,7 +9,7 @@
|
|
|
9
9
|
</p>
|
|
10
10
|
<p align="center">
|
|
11
11
|
<img src="https://img.shields.io/badge/npm-litclaude--ai-cb3837" />
|
|
12
|
-
<img src="https://img.shields.io/badge/version-0.3.
|
|
12
|
+
<img src="https://img.shields.io/badge/version-0.3.40-2ea44f" />
|
|
13
13
|
<img src="https://img.shields.io/badge/Claude%20Code-plugin-blueviolet" />
|
|
14
14
|
<img src="https://img.shields.io/badge/license-MIT-blue" />
|
|
15
15
|
</p>
|
|
@@ -174,18 +174,18 @@ cd /tmp
|
|
|
174
174
|
npx --yes litclaude-ai@latest install
|
|
175
175
|
```
|
|
176
176
|
|
|
177
|
-
`0.3.
|
|
177
|
+
`0.3.40` is the version prepared by this checkout. Registry lookup is authoritative
|
|
178
178
|
for whether that exact version can be installed:
|
|
179
179
|
|
|
180
180
|
```bash
|
|
181
|
-
npm view litclaude-ai@0.3.
|
|
181
|
+
npm view litclaude-ai@0.3.40 version
|
|
182
182
|
```
|
|
183
183
|
|
|
184
|
-
If that lookup returns 0.3.
|
|
184
|
+
If that lookup returns 0.3.40, the exact install is available:
|
|
185
185
|
|
|
186
186
|
```bash
|
|
187
187
|
cd /tmp
|
|
188
|
-
npx --yes litclaude-ai@0.3.
|
|
188
|
+
npx --yes litclaude-ai@0.3.40 install
|
|
189
189
|
```
|
|
190
190
|
|
|
191
191
|
Otherwise, wait for explicit human publication and repeat the lookup before using
|
|
@@ -216,7 +216,7 @@ The installer also sets Claude Code's `statusLine` command to the packaged
|
|
|
216
216
|
LitClaude HUD. A typical no-color render starts like:
|
|
217
217
|
|
|
218
218
|
```text
|
|
219
|
-
[🔥LITCLAUDE v0.3.
|
|
219
|
+
[🔥LITCLAUDE v0.3.40] | O4.8 │ ctx [▎░░] 9%/1000k │ 5h [▏░] 4% ↻2h15m │ 1w [▊░] 35% ↻3d6h │ git main +3 ✓
|
|
220
220
|
```
|
|
221
221
|
|
|
222
222
|
The `↻` suffix is a compact rate-limit reset countdown. It is separated from
|
package/README_ko-KR.md
CHANGED
|
@@ -9,7 +9,7 @@
|
|
|
9
9
|
</p>
|
|
10
10
|
<p align="center">
|
|
11
11
|
<img src="https://img.shields.io/badge/npm-litclaude--ai-cb3837" />
|
|
12
|
-
<img src="https://img.shields.io/badge/version-0.3.
|
|
12
|
+
<img src="https://img.shields.io/badge/version-0.3.40-2ea44f" />
|
|
13
13
|
<img src="https://img.shields.io/badge/Claude%20Code-plugin-blueviolet" />
|
|
14
14
|
<img src="https://img.shields.io/badge/license-MIT-blue" />
|
|
15
15
|
</p>
|
|
@@ -183,18 +183,18 @@ cd /tmp
|
|
|
183
183
|
npx --yes litclaude-ai@latest install
|
|
184
184
|
```
|
|
185
185
|
|
|
186
|
-
`0.3.
|
|
186
|
+
`0.3.40`은 이 checkout에서 준비 중인 version입니다. exact version의 설치 가능
|
|
187
187
|
여부에는 registry 조회 결과를 authoritative한 기준으로 사용합니다.
|
|
188
188
|
|
|
189
189
|
```bash
|
|
190
|
-
npm view litclaude-ai@0.3.
|
|
190
|
+
npm view litclaude-ai@0.3.40 version
|
|
191
191
|
```
|
|
192
192
|
|
|
193
|
-
조회 결과가 0.3.
|
|
193
|
+
조회 결과가 0.3.40이면 exact install을 사용할 수 있습니다.
|
|
194
194
|
|
|
195
195
|
```bash
|
|
196
196
|
cd /tmp
|
|
197
|
-
npx --yes litclaude-ai@0.3.
|
|
197
|
+
npx --yes litclaude-ai@0.3.40 install
|
|
198
198
|
```
|
|
199
199
|
|
|
200
200
|
그렇지 않으면 명시적인 human publication을 기다립니다. 게시 후 조회를 다시 실행해
|
|
@@ -225,7 +225,7 @@ installer는 Claude Code의 `statusLine` command도 packaged LitClaude HUD로
|
|
|
225
225
|
설정합니다. 색상을 제거한 예시는 다음처럼 시작합니다.
|
|
226
226
|
|
|
227
227
|
```text
|
|
228
|
-
[🔥LITCLAUDE v0.3.
|
|
228
|
+
[🔥LITCLAUDE v0.3.40] | O4.8 │ ctx [▎░░] 9%/1000k │ 5h [▏░] 4% ↻2h15m │ 1w [▊░] 35% ↻3d6h │ git main +3 ✓
|
|
229
229
|
```
|
|
230
230
|
|
|
231
231
|
`↻` 표시는 rate-limit reset까지 남은 시간을 짧게 보여주는 countdown입니다.
|
package/RELEASE_CHECKLIST.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
# LitClaude Release Checklist
|
|
2
2
|
|
|
3
|
-
Status: `litclaude-ai@0.3.
|
|
3
|
+
Status: `litclaude-ai@0.3.40` is the current release candidate — exact canonical
|
|
4
4
|
frontend corpus plus Claude-native `autoresearch`, `autoconference`, and `wikify`
|
|
5
5
|
workflow-family integration. It byte-pins the frontend library, legal companions,
|
|
6
6
|
family source closures, and adapters through independent commitments and package
|
|
@@ -45,9 +45,9 @@ side-effect-free, the launcher starts only a separate Claude Code
|
|
|
45
45
|
print/background worker, and the release preserves the Korean polishing
|
|
46
46
|
command, strict multi-agent review pipeline, fidelity guardrails, package
|
|
47
47
|
hygiene checks, native route gates, and safe start-work handoff behavior.
|
|
48
|
-
`package.json` is aligned to `0.3.
|
|
49
|
-
`plugins/litclaude/.claude-plugin/plugin.json` is aligned to `0.3.
|
|
50
|
-
plugin-local MCP server reports `0.3.
|
|
48
|
+
`package.json` is aligned to `0.3.40`,
|
|
49
|
+
`plugins/litclaude/.claude-plugin/plugin.json` is aligned to `0.3.40`, and the
|
|
50
|
+
plugin-local MCP server reports `0.3.40`.
|
|
51
51
|
|
|
52
52
|
This release carries the v0.2.2 Dynamic workflow hardening surfaces:
|
|
53
53
|
`/dynamic-workflow`, `workflow-check --json`, native `/goal` fallback guidance,
|
|
@@ -280,9 +280,9 @@ checkout and from an isolated install of the packed tarball:
|
|
|
280
280
|
Before requesting publication approval, confirm these artifacts from the current
|
|
281
281
|
checkout:
|
|
282
282
|
|
|
283
|
-
- `package.json` version is `0.3.
|
|
284
|
-
- `plugins/litclaude/.claude-plugin/plugin.json` version is `0.3.
|
|
285
|
-
- `plugins/litclaude/bin/litclaude-mcp.js` reports server version `0.3.
|
|
283
|
+
- `package.json` version is `0.3.40`.
|
|
284
|
+
- `plugins/litclaude/.claude-plugin/plugin.json` version is `0.3.40`.
|
|
285
|
+
- `plugins/litclaude/bin/litclaude-mcp.js` reports server version `0.3.40`.
|
|
286
286
|
- Prompt-hook tests cover bundled `SKILL.md` body injection for bare `hyperplan`, `litresearch`, `lit research`, `init-deep`, and explicit leading `$start-work`; diagnostic/copy mentions stay inert while leading natural-language `lit start work` stays BLOCKED.
|
|
287
287
|
- `lit search` and `lit query` route to `/litclaude:litresearch` without activating on slash mentions, code spans, or non-lit prompts.
|
|
288
288
|
- Litresearch web lanes require public API/feed preference, validator-first checks, route traces, prompt-injection quarantine, and honest auth/paywall/private-data stop reasons.
|
package/bin/litclaude-ai.js
CHANGED
|
@@ -100,6 +100,19 @@ const knownExternalLspPlugins = {
|
|
|
100
100
|
};
|
|
101
101
|
const spinnerFrames = ["⠋", "⠙", "⠹", "⠸", "⠼", "⠴", "⠦", "⠧", "⠇", "⠏"];
|
|
102
102
|
const hudAccentThemes = HUD_ACCENT_THEMES;
|
|
103
|
+
const OUTPUT_STYLE_OPTIONS = [
|
|
104
|
+
{ id: "off", label: "None / keep current" },
|
|
105
|
+
{ id: "asd-ste100", label: "ASD-STE100 (English)" },
|
|
106
|
+
{ id: "asd-ste100-ko", label: "ASD-STE100 (한국어)" },
|
|
107
|
+
{ id: "eli5", label: "ELI5 (English)" },
|
|
108
|
+
{ id: "eli5-ko", label: "ELI5 (한국어)" },
|
|
109
|
+
];
|
|
110
|
+
const OUTPUT_STYLE_NAMES = {
|
|
111
|
+
"asd-ste100": "LitClaude — ASD-STE100",
|
|
112
|
+
"asd-ste100-ko": "LitClaude — ASD-STE100 (한국어)",
|
|
113
|
+
"eli5": "LitClaude — ELI5",
|
|
114
|
+
"eli5-ko": "LitClaude — ELI5 (한국어)",
|
|
115
|
+
};
|
|
103
116
|
const permissionModes = new Set(["safe", "balanced", "yolo"]);
|
|
104
117
|
const permissionOwnershipDescription =
|
|
105
118
|
"preserves pre-existing entries; removes only LitClaude-inserted entries on mode change or uninstall";
|
|
@@ -351,6 +364,50 @@ const chooseHudAccent = async (existingAccent = "cyan") => {
|
|
|
351
364
|
}
|
|
352
365
|
};
|
|
353
366
|
|
|
367
|
+
const normalizeOutputStyle = (value) => {
|
|
368
|
+
if (!value) return "off";
|
|
369
|
+
const v = String(value).trim().toLowerCase();
|
|
370
|
+
const match = OUTPUT_STYLE_OPTIONS.find((opt) => opt.id === v);
|
|
371
|
+
return match ? match.id : "off";
|
|
372
|
+
};
|
|
373
|
+
|
|
374
|
+
const shouldPromptOutputStyle = () => {
|
|
375
|
+
if (process.env.LITCLAUDE_OUTPUT_STYLE_PROMPT === "1") return true;
|
|
376
|
+
if (process.env.LITCLAUDE_OUTPUT_STYLE_PROMPT === "0") return false;
|
|
377
|
+
return Boolean(process.stdin.isTTY && process.stdout.isTTY && !process.env.CI);
|
|
378
|
+
};
|
|
379
|
+
|
|
380
|
+
const chooseOutputStyle = async (existingChoice = "off") => {
|
|
381
|
+
const preserved = normalizeOutputStyle(existingChoice);
|
|
382
|
+
const envStyle = process.env.LITCLAUDE_OUTPUT_STYLE;
|
|
383
|
+
if (envStyle && process.env.LITCLAUDE_OUTPUT_STYLE_PROMPT !== "1") {
|
|
384
|
+
return normalizeOutputStyle(envStyle);
|
|
385
|
+
}
|
|
386
|
+
if (!shouldPromptOutputStyle()) return envStyle ? normalizeOutputStyle(envStyle) : preserved;
|
|
387
|
+
|
|
388
|
+
process.stdout.write("Choose LitClaude output style:\n");
|
|
389
|
+
for (const [index, option] of OUTPUT_STYLE_OPTIONS.entries()) {
|
|
390
|
+
process.stdout.write(` ${index}) ${option.label}\n`);
|
|
391
|
+
}
|
|
392
|
+
|
|
393
|
+
const preservedLabel = OUTPUT_STYLE_OPTIONS.find((opt) => opt.id === preserved)?.label ?? preserved;
|
|
394
|
+
const rl = createInterface({ input: process.stdin, output: process.stdout });
|
|
395
|
+
try {
|
|
396
|
+
const answer = await rl.question(`Style 0-4 [${preservedLabel}]: `);
|
|
397
|
+
const trimmed = answer.trim();
|
|
398
|
+
const byIndex = OUTPUT_STYLE_OPTIONS[Number.parseInt(trimmed, 10)]?.id;
|
|
399
|
+
const selected = byIndex !== undefined ? byIndex : (trimmed ? normalizeOutputStyle(trimmed) : preserved);
|
|
400
|
+
const isExplicitOff = trimmed.toLowerCase() === "off" || trimmed.toLowerCase() === "none";
|
|
401
|
+
if (trimmed && byIndex === undefined && selected === "off" && !isExplicitOff) {
|
|
402
|
+
process.stdout.write("OUTPUT_STYLE_WARNING: unknown style; using 'off'\n");
|
|
403
|
+
}
|
|
404
|
+
process.stdout.write(`OUTPUT_STYLE_SELECTED: ${selected}\n`);
|
|
405
|
+
return selected;
|
|
406
|
+
} finally {
|
|
407
|
+
rl.close();
|
|
408
|
+
}
|
|
409
|
+
};
|
|
410
|
+
|
|
354
411
|
const hudCommandForPlugin = (pluginPath, accent = "cyan") =>
|
|
355
412
|
`LITCLAUDE_HUD_ACCENT=${normalizeHudAccent(accent)} node "${join(pluginPath, "bin", "litclaude-hud.js")}"`;
|
|
356
413
|
|
|
@@ -458,6 +515,30 @@ const installHudStatusLine = (pluginPath, home = claudeHome(), accent = "cyan")
|
|
|
458
515
|
writeClaudeSettings(settings, home);
|
|
459
516
|
};
|
|
460
517
|
|
|
518
|
+
const installOutputStyle = (choice, home = claudeHome()) => {
|
|
519
|
+
if (choice === "off") return;
|
|
520
|
+
const writtenName = OUTPUT_STYLE_NAMES[choice] ?? null;
|
|
521
|
+
if (!writtenName) return;
|
|
522
|
+
|
|
523
|
+
const settings = readClaudeSettings(home);
|
|
524
|
+
const existingLitClaude = settings[litClaudeSettingsKey] ?? {};
|
|
525
|
+
const previousWrittenName = existingLitClaude.outputStyleWrittenName ?? null;
|
|
526
|
+
|
|
527
|
+
const currentOutputStyle = settings.outputStyle;
|
|
528
|
+
if (currentOutputStyle === undefined || currentOutputStyle === previousWrittenName) {
|
|
529
|
+
settings.outputStyle = writtenName;
|
|
530
|
+
} else {
|
|
531
|
+
process.stdout.write(`Note: outputStyle was changed outside LitClaude (currently "${currentOutputStyle}"). Switch it via /config if you want LitClaude's style.\n`);
|
|
532
|
+
}
|
|
533
|
+
|
|
534
|
+
settings[litClaudeSettingsKey] = {
|
|
535
|
+
...existingLitClaude,
|
|
536
|
+
outputStyleChoice: choice,
|
|
537
|
+
outputStyleWrittenName: writtenName,
|
|
538
|
+
};
|
|
539
|
+
writeClaudeSettings(settings, home);
|
|
540
|
+
};
|
|
541
|
+
|
|
461
542
|
const uninstallHudStatusLine = (home = claudeHome()) => {
|
|
462
543
|
const settingsPath = claudeSettingsPath(home);
|
|
463
544
|
if (!existsSync(settingsPath)) return;
|
|
@@ -840,6 +921,7 @@ const install = async ({ dryRun, rest }) => {
|
|
|
840
921
|
const existingLitClaude = readClaudeSettings(home)[litClaudeSettingsKey] ?? {};
|
|
841
922
|
const { permissionMode, permissionExplicit } = parseInstallOptions(rest, existingLitClaude.permissionMode);
|
|
842
923
|
const existingHudAccent = normalizeHudAccent(existingLitClaude.hudAccent);
|
|
924
|
+
const existingOutputStyle = existingLitClaude.outputStyleChoice ?? "off";
|
|
843
925
|
const requestedHudAccent = process.env.LITCLAUDE_HUD_ACCENT
|
|
844
926
|
? normalizeHudAccent(process.env.LITCLAUDE_HUD_ACCENT)
|
|
845
927
|
: existingHudAccent;
|
|
@@ -868,6 +950,7 @@ const install = async ({ dryRun, rest }) => {
|
|
|
868
950
|
}
|
|
869
951
|
|
|
870
952
|
const hudAccent = await chooseHudAccent(existingHudAccent);
|
|
953
|
+
const outputStyleChoice = await chooseOutputStyle(existingOutputStyle);
|
|
871
954
|
progress.prepare({ claudeTarget: home, litTarget: litClaudeHome });
|
|
872
955
|
await progress.run(
|
|
873
956
|
{
|
|
@@ -901,7 +984,10 @@ const install = async ({ dryRun, rest }) => {
|
|
|
901
984
|
purpose: "Enable the plugin and HUD, then update global Claude settings permissions.allow/deny",
|
|
902
985
|
target: claudeSettingsPath(home),
|
|
903
986
|
},
|
|
904
|
-
() =>
|
|
987
|
+
() => {
|
|
988
|
+
registerClaudePlugin(targetPlugin, localMarketplace, home, hudAccent, permissionMode);
|
|
989
|
+
installOutputStyle(outputStyleChoice, home);
|
|
990
|
+
},
|
|
905
991
|
);
|
|
906
992
|
|
|
907
993
|
const litRoot = versionRoot(litClaudeHome);
|
package/package.json
CHANGED
|
@@ -0,0 +1,46 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: "LitClaude — ASD-STE100"
|
|
3
|
+
description: Write responses in ASD-STE100 Simplified Technical English — short sentences, active voice, controlled vocabulary, one instruction per step.
|
|
4
|
+
keep-coding-instructions: true
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
Write all prose output in ASD-STE100 Simplified Technical English (STE). This applies to
|
|
8
|
+
explanations, summaries, and instructions you give the user — not to code, code comments, file
|
|
9
|
+
contents, or quoted material, which follow their own conventions.
|
|
10
|
+
|
|
11
|
+
## Sentence construction
|
|
12
|
+
|
|
13
|
+
- One idea per sentence. Do not join unrelated ideas with "and" or "which."
|
|
14
|
+
- Keep sentences short: 20 words or fewer for descriptive sentences, 15-20 words for instructions.
|
|
15
|
+
- Use active voice. Name the agent that does the action.
|
|
16
|
+
- Write "Run the migration script." not "The migration script should be run."
|
|
17
|
+
- Use simple verb tenses only:
|
|
18
|
+
- Simple present for facts and general truths ("The function returns null.")
|
|
19
|
+
- Simple past for completed actions ("The build failed.")
|
|
20
|
+
- Simple future for future events ("The test will fail.")
|
|
21
|
+
- Avoid continuous, perfect, and perfect-continuous tenses ("has been running," "will have completed").
|
|
22
|
+
- Always use articles ("a," "an," "the"). Do not drop them for brevity.
|
|
23
|
+
- Limit noun clusters to 3 nouns in a row. Break up long noun strings with prepositions.
|
|
24
|
+
- Write "the timeout value for the connection pool" not "the connection pool timeout value."
|
|
25
|
+
- Avoid "-ing" forms as nouns or adjectives where a plain verb form exists.
|
|
26
|
+
- Write "to configure the server" not "for configuring the server."
|
|
27
|
+
|
|
28
|
+
## Vocabulary
|
|
29
|
+
|
|
30
|
+
- One word, one meaning. Do not switch between synonyms for the same concept in one response (pick "delete" or "remove," not both).
|
|
31
|
+
- Avoid jargon, idioms, and figurative language. State things literally.
|
|
32
|
+
- Define a technical term the first time you use it, in simple terms.
|
|
33
|
+
- Prefer common, concrete words over abstract or formal ones.
|
|
34
|
+
|
|
35
|
+
## Instructions and procedures
|
|
36
|
+
|
|
37
|
+
- Write instructions as numbered steps, in imperative mood, present tense.
|
|
38
|
+
- One action per step. Split compound actions into separate steps.
|
|
39
|
+
- Put a step's condition before its action: "If the file exists, delete it." not "Delete the file if it exists."
|
|
40
|
+
- Flag risk or required attention with an explicit label before the relevant step: `WARNING:`, `CAUTION:`, or `NOTE:`.
|
|
41
|
+
|
|
42
|
+
## Structure
|
|
43
|
+
|
|
44
|
+
- Prefer short paragraphs, numbered steps, and bullet lists over dense prose.
|
|
45
|
+
- State the main point first, then supporting detail.
|
|
46
|
+
- If a rule above conflicts with brevity, choose clarity and unambiguous meaning.
|
|
@@ -0,0 +1,47 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: "LitClaude — ASD-STE100 (한국어)"
|
|
3
|
+
description: 짧은 문장, 능동태, 통제된 어휘, 한 단계에 한 지시로 쓰는 쉬운 기술 문장체.
|
|
4
|
+
keep-coding-instructions: true
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
<!-- Principle-adapted from the English version, not translated. Do not re-translate. -->
|
|
8
|
+
|
|
9
|
+
설명, 요약, 사용자에게 주는 지시는 모두 쉬운 기술 문장체로 쓰십시오. 코드, 코드 주석, 파일 내용,
|
|
10
|
+
인용문에는 이 규칙을 적용하지 마십시오. 그것들은 각자의 규칙을 따릅니다.
|
|
11
|
+
|
|
12
|
+
## 문장 구성
|
|
13
|
+
|
|
14
|
+
- 한 문장에 한 가지 생각만 담으십시오. 관계없는 내용을 "그리고"나 "그런데"로 잇지 마십시오.
|
|
15
|
+
- 문장을 짧게 쓰십시오. 설명문은 40자 안팎, 지시문은 30자 안팎을 넘기지 마십시오.
|
|
16
|
+
- 능동태를 쓰십시오. 행동의 주체를 밝히십시오.
|
|
17
|
+
- "마이그레이션 스크립트를 실행하십시오."라고 쓰십시오. "마이그레이션 스크립트가 실행되어야
|
|
18
|
+
합니다."라고 쓰지 마십시오.
|
|
19
|
+
- 시제는 단순하게 쓰십시오.
|
|
20
|
+
- 사실이나 일반 원리는 현재형: "이 함수는 null을 반환합니다."
|
|
21
|
+
- 끝난 일은 과거형: "빌드가 실패했습니다."
|
|
22
|
+
- 앞으로 일어날 일은 미래형: "이 테스트는 실패할 것입니다."
|
|
23
|
+
- "~하고 있었습니다", "~해 왔습니다" 같은 복잡한 시제는 피하십시오.
|
|
24
|
+
- 긴 명사 나열을 피하십시오. 조사로 끊어 쓰십시오.
|
|
25
|
+
- "연결 풀의 타임아웃 값"이라고 쓰십시오. "연결 풀 타임아웃 값"처럼 명사를 겹쳐 쓰지 마십시오.
|
|
26
|
+
|
|
27
|
+
## 어휘
|
|
28
|
+
|
|
29
|
+
- 한 단어에 한 뜻만 쓰십시오. 같은 개념을 한 답변 안에서 다른 말로 바꿔 쓰지 마십시오. ("삭제"면
|
|
30
|
+
"삭제"로 끝까지 쓰고, "제거"와 섞지 마십시오.)
|
|
31
|
+
- 전문 용어, 관용구, 비유적 표현을 피하십시오. 있는 그대로 말하십시오.
|
|
32
|
+
- 전문 용어는 처음 쓸 때 쉬운 말로 뜻을 설명하십시오.
|
|
33
|
+
- 어렵고 추상적인 말보다 쉽고 구체적인 말을 쓰십시오.
|
|
34
|
+
|
|
35
|
+
## 지시와 절차
|
|
36
|
+
|
|
37
|
+
- 지시는 번호가 붙은 단계로, 명령형·현재형으로 쓰십시오.
|
|
38
|
+
- 한 단계에 한 가지 행동만 넣으십시오. 여러 행동은 단계를 나누십시오.
|
|
39
|
+
- 조건을 행동보다 먼저 쓰십시오. "파일이 있으면 삭제하십시오."라고 쓰십시오. "파일을 삭제하십시오,
|
|
40
|
+
있다면."이라고 쓰지 마십시오.
|
|
41
|
+
- 위험하거나 주의가 필요한 단계 앞에는 `경고:`, `주의:`, `참고:`를 명확히 표시하십시오.
|
|
42
|
+
|
|
43
|
+
## 구조
|
|
44
|
+
|
|
45
|
+
- 빽빽한 글보다 짧은 문단, 번호 목록, 글머리표를 쓰십시오.
|
|
46
|
+
- 핵심을 먼저 말하고, 세부 내용을 뒤에 쓰십시오.
|
|
47
|
+
- 위 규칙과 간결함이 부딪히면, 명확하고 오해 없는 뜻을 선택하십시오.
|
|
@@ -0,0 +1,17 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: "LitClaude — ELI5"
|
|
3
|
+
description: keep it simple pls
|
|
4
|
+
keep-coding-instructions: true
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
It's been a long day and my brain is fried, talk to me like I'm 5.
|
|
8
|
+
|
|
9
|
+
Small words, short sentences, short paragraphs. If you have to use
|
|
10
|
+
a big word, explain it right after. Only return what's actually necessary.
|
|
11
|
+
|
|
12
|
+
Just tell me what you did, did it work, what do I do now.
|
|
13
|
+
|
|
14
|
+
If I have to decide something: 2 options max, the context I need to pick fast,
|
|
15
|
+
and which one you'd go with.
|
|
16
|
+
|
|
17
|
+
Keep paths and commands exact. I have no brain cells left for the rest.
|
|
@@ -0,0 +1,19 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: "LitClaude — ELI5 (한국어)"
|
|
3
|
+
description: 쉽게, 짧게, 필요한 것만.
|
|
4
|
+
keep-coding-instructions: true
|
|
5
|
+
---
|
|
6
|
+
|
|
7
|
+
<!-- Principle-adapted from the English version, not translated. Do not re-translate. -->
|
|
8
|
+
|
|
9
|
+
오늘 하루가 길었고 머리가 지쳤어요. 다섯 살짜리한테 설명하듯 말해주세요.
|
|
10
|
+
|
|
11
|
+
짧은 단어, 짧은 문장, 짧은 문단을 쓰세요. 어려운 단어를 써야 한다면 바로 뒤에 쉽게 풀어서
|
|
12
|
+
설명하세요. 정말 필요한 내용만 답하세요.
|
|
13
|
+
|
|
14
|
+
무엇을 했는지, 잘 됐는지, 다음에 뭘 해야 하는지만 말해주세요.
|
|
15
|
+
|
|
16
|
+
제가 뭔가 결정해야 한다면: 선택지는 최대 2개, 빠르게 고르는 데 필요한 맥락만, 그리고 당신이라면
|
|
17
|
+
어떤 걸 고를지도 알려주세요.
|
|
18
|
+
|
|
19
|
+
경로와 명령어는 정확하게 그대로 써주세요. 그 외의 것까지 신경 쓸 여유가 없어요.
|