thai-config
Skill0 starsChanged 22 days agoMITView on GitHub
Show or change thai-native settings, such as default register, your particle (ครับ/ค่ะ), tech-term policy, numerals, slang, lint hooks and per-term overrides. Saves to your user config or to the project's .thai-native/config.json for the whole team.
In this repository6 other files it gives its agents.
- exports/AGENTS.mdAGENTS.md
- thai-learnSkill
- thai-polishSkill
- thai-termSkill
- thai-translateSkill
- thai-writingSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 35
Add --project to save in .thai-native/config.json at the project root (commit it to share with the team).
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
thai-config
Script: python3 "${CLAUDE_SKILL_DIR}/../thai-writing/scripts/thai_native.py" --project-dir "${CLAUDE_PROJECT_DIR}" config
- With arguments (
$ARGUMENTS): run the script with them and report the result in one or two Thai lines. - Without arguments: run
config show, then offer a quick setup. Ask, in one round, (1) particle: ครับ / ค่ะ / ไม่ใช้คำลงท้าย / ดูจากข้อความ, (2) default register: auto / team / business / formal / casual / social, (3) tech terms: ตามคลังคำ (auto) / คงอังกฤษ / ทับศัพท์ / ไทย (อังกฤษ) / แปลไทย, (4) lint after writing files: warn / off. Save all answers with one command,config set particle=… register=… term_policy=… file_hook=…, and show the final config. Until these four are set, the plugin's prompt hook asks the same questions once per session before a Thai task (setup_prompt offstops it).
| key | values | meaning |
register | auto, formal, business, team, casual, social | default register (auto = by genre) |
particle | auto, krap, ka, none | ครับ / ค่ะ / none; auto mirrors the user |
term_policy | auto, keep_en, translit, th_en, translate | override every glossary term at once |
numerals | arabic, thai | digits |
year | auto, be, ce | พ.ศ. or ค.ศ. in dates |
slang | off, light, on | slang in casual/social registers |
emoji | auto, off, on | emoji in chat |
self_check | auto, lint, checklist | when to run the lint |
file_hook | warn, off | lint Markdown/text files after Write/Edit |
prompt_hook | auto, always, off | add how the glossary writes the tech terms in your message (auto = only when the message is about Thai) |
prompt_cards | off, on | also push short reference cards for the task with the glossary note (genre rules once per session, register, translation patterns). Off by default: in the 0.2.0 A/B, answers with pushed cards lost more often than without |
stop_hook | auto, off | if Claude's final reply has a certain Thai error (ครับ/ค่ะ with a slash, misspelled particle), ask once for a corrected reply |
setup_prompt | auto, off | until particle, register, term_policy and file_hook are set, ask for them once per session before a Thai task |
post_hook | off, warn, block-once | lint Thai text before Jira/Confluence/Slack/Linear/Notion MCP tools post it |
learn | off, on | thai-learn: keep local copies of Thai files Claude writes so it can learn from your edits (user config only; a project file cannot turn it on) |
term | "<term> <policy> [thai]" | personal override for one term |
Add --project to save in .thai-native/config.json at the project root (commit it to share with the team). Claude Code users can also set register, particle, term policy and the file hook in /plugin configure thai-native@thai-native; the config file wins when both are set.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.