K
KalimIzy/persian-colloquial-skill
Convert formal Persian to Tehran colloquial Persian, and check the register of Persian text.
updated 2026-11-18
View on GitHub ↗Direct link · no redirect page
After install: how to tell it works
- What you see after install — The command exits 0 and the terminal shows no "command not found" or "permission denied".
- One-line self-check — Run the smallest usage example from that repository's README. If the output matches what the README describes, it is working.
- If that does not work — This project has no troubleshooting notes. Search its issues for "install".
The install command comes from the repository; these checks describe observable results only and do not replace the project's own docs.
- Star count
- 0 ★ · KalimIzy/persian-colloquial-skill · read from the GitHub search API · (checked 2026-11-19 20:00:48)
- Source
- read directly from its GitHub page · type: open-source project
- Harvest history
- first seen 2026-11-19 · last seen 2026-11-20 · seen 2× · listed by 4 sources
- Score
- score 403 · 4 sources · repo
- Install
- see Install & config above
FAQ (answers come from our records)
Who is it for?
We filed it under these tags: Agent skills. The tags come from the entry itself, not a taxonomy we imposed.
Who maintains it?
repository
KalimIzy/persian-colloquial-skill;first catalogued 2026-11-19;last seen 2026-11-20。There is no official certification or quality rating here — we record sources and dates only.README excerpt (English, untranslated) · Read the full README on GitHub →
# Persian Colloquial Skill
Excerpted from the repository README (abridged)
Convert formal Persian into widely understood Tehran colloquial Persian, and check the register of any Persian text. Built for AI agents, useful for humans too.
Formal Persian and spoken Persian are not the same language with different spelling. Verbs contract (میروم → میرم), the copula attaches (خوب است → خوبه), the object marker shrinks (را → رو), vowels shift (خانه → خونه), and whole words change (اگر → اگه, برای → واسه). This skill teaches an agent all of it, with rules grounded in real data instead of vibes.
## Quick example
Formal:
> سلام، چطور هستی؟ من دیروز به دانشگاه رفتم و استاد گفت که امتحان به هفته آینده موکول شده است.
Colloquial:
> سلام، چطوری؟ دیروز رفتم دانشگاه، استاد گفت امتحان افتاده هفته دیگه.
## Choosing the register
The skill does not only convert — it decides which register fits. Five levels, from very formal to full colloquial, and a decision engine that combines the incoming message (mirroring), a stored contact profile, and an audience × channel matrix:
bash
python skills/persian-colloquial/scripts/suggest_register.py --incoming "سلام چطوری؟ خوبی؟"
python skills/persian-colloquial/scripts/suggest_register.py --audience teacher --channel email --topic admin
python skills/persian-colloquial/scripts/suggest_register.py --contact مامان --profile contacts.yaml
python skills/persian-colloquial/scripts/suggest_register.py --prompt "یه پروپوزال واسه مشتری بنویس" --deliverable proposal
The deliverable type beats the request tone (a casual request for a proposal still needs a formal proposal); the live message beats the stored profile, the profile beats the matrix, and level 3 (semi-formal) is the safe default when nothing is known. The user's own request (--prompt) is never mirrored: asking casually for a formal text must not drag the reply colloquial. See references/register-decision.md for the full rules and references/register-levels.md for what each level sounds like.
## What is inside
| File | What it holds |
| skills/persian-colloquial/SKILL.md | The agent skill: conversion rules, register consistency, politeness, checking |
| skills/persian-colloquial/references/register-levels.md | The five register levels, with rules and one message at all five |
| skills/persian-colloquial/references/register-decision.md | How the register is decided: precedence, matrix, confidence |
| skills/persian-colloquial/references/orthography.md | ZWNJ, Persian characters, digits, punctuation, هکسره, tanvin |
| skills/persian-colloquial/references/ai-tells.md | Bureaucratic and AI tells in Persian, with fixes |
| skills/persian-colloquial/scripts/suggest_register.py | Decision engine: mirror, profile, matrix → level + confidence |
| skills/persian-colloquial/examples/contacts.example.yaml | Example contacts profile (copy to contacts.yaml, gitignored) |
| skills/persian-colloquial/references/verb-tables.md | Full formal vs colloquial tables for the top 40 verbs, hand-checked |
| skills/persian-colloquial/references/verb-tables-full.md | All 341 verbs with stems and colloquial present tense |
| skills/persian-colloquial/references/word-pairs.md | 195+ colloquial to formal word pairs, grouped by pattern |
Formal Persian and spoken Persian are not the same language with different spelling. Verbs contract (میروم → میرم), the copula attaches (خوب است → خوبه), the object marker shrinks (را → رو), vowels shift (خانه → خونه), and whole words change (اگر → اگه, برای → واسه). This skill teaches an agent all of it, with rules grounded in real data instead of vibes.
## Quick example
Formal:
> سلام، چطور هستی؟ من دیروز به دانشگاه رفتم و استاد گفت که امتحان به هفته آینده موکول شده است.
Colloquial:
> سلام، چطوری؟ دیروز رفتم دانشگاه، استاد گفت امتحان افتاده هفته دیگه.
## Choosing the register
The skill does not only convert — it decides which register fits. Five levels, from very formal to full colloquial, and a decision engine that combines the incoming message (mirroring), a stored contact profile, and an audience × channel matrix:
bash
python skills/persian-colloquial/scripts/suggest_register.py --incoming "سلام چطوری؟ خوبی؟"
python skills/persian-colloquial/scripts/suggest_register.py --audience teacher --channel email --topic admin
python skills/persian-colloquial/scripts/suggest_register.py --contact مامان --profile contacts.yaml
python skills/persian-colloquial/scripts/suggest_register.py --prompt "یه پروپوزال واسه مشتری بنویس" --deliverable proposal
The deliverable type beats the request tone (a casual request for a proposal still needs a formal proposal); the live message beats the stored profile, the profile beats the matrix, and level 3 (semi-formal) is the safe default when nothing is known. The user's own request (--prompt) is never mirrored: asking casually for a formal text must not drag the reply colloquial. See references/register-decision.md for the full rules and references/register-levels.md for what each level sounds like.
## What is inside
| File | What it holds |
| skills/persian-colloquial/SKILL.md | The agent skill: conversion rules, register consistency, politeness, checking |
| skills/persian-colloquial/references/register-levels.md | The five register levels, with rules and one message at all five |
| skills/persian-colloquial/references/register-decision.md | How the register is decided: precedence, matrix, confidence |
| skills/persian-colloquial/references/orthography.md | ZWNJ, Persian characters, digits, punctuation, هکسره, tanvin |
| skills/persian-colloquial/references/ai-tells.md | Bureaucratic and AI tells in Persian, with fixes |
| skills/persian-colloquial/scripts/suggest_register.py | Decision engine: mirror, profile, matrix → level + confidence |
| skills/persian-colloquial/examples/contacts.example.yaml | Example contacts profile (copy to contacts.yaml, gitignored) |
| skills/persian-colloquial/references/verb-tables.md | Full formal vs colloquial tables for the top 40 verbs, hand-checked |
| skills/persian-colloquial/references/verb-tables-full.md | All 341 verbs with stems and colloquial present tense |
| skills/persian-colloquial/references/word-pairs.md | 195+ colloquial to formal word pairs, grouped by pattern |
Listed badge (optional)
Copy this line into the README if you want the listing mark. Optional, and it does not affect being listed.
[](https://anclum.com/en/s/kalimizy-persian-colloquial-skill/)
- Source
- github · https://github.com/KalimIzy/persian-colloquial-skill
- Browse by field
- Agent skills
- Listed by
- 4 independent sources
- added
- today
More entries
Filed under
Every tag opens the full list for that filter. Counts are the live catalogue figures.