Skip to main content

مَنْهَج (Manhaj) — أكاديمية الإتقان العلمي العربي المفتوحة

Open-Source Arabic Scientific Mastery Academy & Zero-Custody BYOK Runtime
Part of the Maqāyīs (مَقاييس) Computational Arabic Linguistics Ecosystem · Live Academy (maqayis.org/manhaj/) · GitHub (maqayis/manhaj) · Codeberg Mirror

License: Code Apache-2.0 License: Curriculum CC BY 4.0


نبذة بالعربية | Overview

مَنْهَج (Manhaj) منصّةُ إتقانٍ علميّ مفتوحةُ المصدر، صُمِّمت لتدريس اللسانيات الحاسوبية العربية ورياضياتها التأسيسية بالعربية الفصحى الرصينة، مرافِقةً لكتاب «هندسة اللسان العربي: الفيزياء الصوتية، الجبر الجذري، والتركيب الدلالي» (الإصدار 2.0).

Unlike passive lecture portals, Manhaj models curriculum as a validated prerequisite Directed Acyclic Graph (DAG) with deterministic pedagogical gates:

  1. Two Interlinked Tracks (23 Rigorous Arabic Units, 305 Standardized Glossary Terms, 107 Course KB Documents):
    • cl-nlp-handasat-allisan (19 Core Units): Triliteral Root & Morphological Mīzān, Sībawayh Vocal-Tract Kinematics, Blind Delexicalization & Leakage-Free Evaluation, Spherical Semantic Manifold & CSLS, Zero-LLM Spectral Graph & Nyström, Stasheff Associahedron & Tree-Cut Subtraction, Riemannian Tangent Space & Geodesic Maps, Grassmannian Subspaces & Jordan Angles, Conjugate vMF & Epistemic Concentration, Positional Coherence Hierarchy, Sub-Root Composition & Mendeleev Table, Two-Scale Merge & Tree-Cut Hierarchy, Riemannian Morphological Operators, Vocal-Tract Kinematics & Qur'anic Two-Regime Law, Differential Geometry of Qur'anic Furūq, Zero-LLM Proof & Empirical-Bayes vMF, Grassmannian Fiber Bundles & Jāmid Excision, Riemannian Metaphor Transport, and Sub-Morphemic Lexical Compression (ʾĪjāz).
    • math-handasat-allisan (4 Active L1 Mathematical Foundation Units): Set Theory & Inclusion–Exclusion (M1.1), Combinatorics, Equivalence Quotients & Stasheff Trees (M1.2), Vector Spaces, Inner Products & Discrete Kinematics (M2.1), and Probability, Bayes, Expectation & Logarithms (M5.1).
  2. 4-Pillar Explain-Back Mastery Defense (دفاع الفهم):
    • Learners unlock downstream nodes on the Knowledge Fringe (حافة المعرفة) by defending each concept in their own Arabic prose across four explicit criteria: Definition (P-Def), Mathematical Formulation (P-Math), Empirical Evidence (P-Emp), and Computational Realization (P-Code), with a deterministic 25-word floor before any LLM call is invoked.
  3. Zero-Custody Browser Runtime ($0/mo Static Hosting + BYOK OpenRouter PKCE):
    • Runs 100% statically on Cloudflare Pages (portal/public/manhaj/ at https://maqayis.org/manhaj/) with local-first IndexedDB persistence, morphological root search, grapheme-cluster-safe Arabic text anchoring, a Guided-Mode Solution ACL Firewall (403 until an exercise is explicitly revealed in the active unit), and RFC 7636 OAuth PKCE (OpenRouter Bring-Your-Own-Key) where credentials never touch Maqāyīs servers.
  4. Double-Scrubbed Opt-In Research Telemetry (Misbar Spine):
    • Strictly local-only (Tier 0) by default; optional anonymous research traces (POST /api/v1/traces) apply client-side + edge regex redaction ([REDACTED_API_KEY], [REDACTED_EMAIL], [REDACTED_PHONE]) and 1-click cryptographic right-to-erasure (DELETE /api/v1/traces).

Repository Structure

Path Description License
concepts/<concept-id>/ Validated concept dossiers (dossier.md) + published Arabic lessons (lessons/*.md) CC BY 4.0
graph/ Prerequisite concept graph (edges.tsv, tracks/*.tsv, map.md) CC BY 4.0
fields/ & fihris/ Field charters (cl-nlp.md), bilingual terminology glossaries, and bibliographic registry CC BY 4.0
corpus/ Grounded Arabic corpus slices and deterministic terminology extraction tools Apache-2.0 / CC BY 4.0
pipeline/validate.py Deterministic 14-gate curriculum, terminology, math, and provenance validator Apache-2.0
learner/kb/ Arabic grapheme-safe document parser, morphological root search, and verification suites Apache-2.0
learner/ui/ RTL Arabic single-page learner application, BYOK ai-engine.js, public-runtime.js, and local serve.py Apache-2.0
deploy/build_public_site.py Static SEO & subpath-capable compiler (333 RTL HTML pages + static-api/ + llms.txt + sitemap.xml) Apache-2.0
deploy/verify_public_deployment.py End-to-end 5-suite verification harness (Python + Node 22 + SQLite Misbar bridge) Apache-2.0

Quickstart

1. Build the Static Public Academy (/manhaj Subpath or Root Domain)

# Build integrated under Maqāyīs Portal (/manhaj subpath + shared /api/v1/traces):
uv run --no-project python deploy/build_public_site.py \
  --out dist \
  --base-url https://maqayis.org/manhaj \
  --trace-endpoint /api/v1/traces \
  --hub-url /

# Serve locally to inspect static bundle:
python3 -m http.server 8080 --directory dist

2. Run the Local Interactive Learner Server (Offline / Dev Mode)

uv run --no-project python learner/ui/serve.py --port 8765 --learner student

3. Run All Curriculum & Public Deployment Verification Gates

# Validate curriculum lessons across both tracks:
uv run --no-project python pipeline/validate.py --concept triliteral-root-and-mizan
uv run --no-project python pipeline/validate.py --concept set-theory-and-inclusion-exclusion --track math-handasat-allisan

# Run the full 5-suite end-to-end public deployment verification (Python + Node 22):
uv run --no-project python deploy/verify_public_deployment.py

Citation & Licensing

  • Platform, Compiler, & Runtime Code: Licensed under the Apache License 2.0 (LICENSE).
  • Curriculum, Lessons, Concept Dossiers, & Glossaries: Licensed under Creative Commons Attribution 4.0 International (CC BY 4.0) (NOTICE).
  • Author: Salah Abdo (صلاح عبده) — <Salah.Abdo.Tech@gmail.com> (Maqāyīs Organization).
@software{abdo2026manhaj,
  author    = {Abdo, Salah},
  title     = {{Manhaj}: Open Arabic Scientific Mastery Academy and Zero-Custody {BYOK} Runtime},
  year      = {2026},
  version   = {1.0.0},
  publisher = {Maq\={a}y\={\i}s},
  url       = {https://maqayis.org/manhaj/},
  note      = {Code: Apache-2.0; Curriculum: CC BY 4.0}
}

Metadata

Release files for manhaj 2.0.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for manhaj 2.0.0
File Size Uploaded Provenance Actions
manhaj-2.0.0.tar.gz 6.2 MB With provenance Details for manhaj-2.0.0.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for manhaj 2.0.0
File Interpreter ABI Platform Actions
manhaj-2.0.0-py3-none-any.whl Python 3 none any Details for manhaj-2.0.0-py3-none-any.whl

Total release size: 6.2 MB

Release files / manhaj-2.0.0.tar.gz

Download URL manhaj-2.0.0.tar.gz
Size 6.2 MB
Tags Source
SHA-256 checksum
How to use checksums
7f217020425489063aaa0b1876509715d664e3b1b710413b8602480ed56ca3ae
BLAKE2b-256 checksum
How to use checksums
44cf1b0f162f30db3a7968663707be799bfb0ffca69a7a241a4eef78efa9cd08
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 8, 2026.

Transparency log

Release files / manhaj-2.0.0-py3-none-any.whl

Download URL manhaj-2.0.0-py3-none-any.whl
Size 9.6 kB
Tags Python 3
SHA-256 checksum
How to use checksums
a551897bd01e3c1e7f1dc162571ac5bddc3b3a66c4ac0f1a295dfc81e9fddb3d
BLAKE2b-256 checksum
How to use checksums
fb2f473b7a67e4ded2e73a51636b56ef6924219aa1598f646eaf7766bdfe40de
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 8, 2026.

Transparency log

Security for manhaj 2.0.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Full provenance

Both files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for manhaj-2.0.0.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 8, 2026.

Transparency log

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

2.0.0 This release

2 release files Full provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page