Open source platform for researching and building trustworthy, Sharia-compliant enterprise products.
This repository also configures its own agents. See what ose-public tells them →
Agent Claude Code
Re-validates each repository rules finding against current files, applies only high-confidence repairs through Rules Propagation with the higher governance level as authority, and hands every open judgement to the rule's owner.
Agent Claude Code
Creates repository rules and conventions in repo-governance/ directories. Documents standards, patterns, and quality requirements.
Agent Claude Code
Creates LinkedIn posts in social-media-posts/linkedin/ from completed origin/main updates across the ose-public and private-sibling repos, enforcing the 3,000-character body limit (from the "OPEN SHARIA ENTERPRISE" line down) in a professional, engaging tone.
Agent Claude Code
Audits explicitly listed specification folders for index quality, scenario format, cross-folder consistency, architecture views, references, and implementation alignment, and returns rated findings without modifying anything.
Agent Claude Code
Applies specification checker findings inside the folders that audit covered, after re-validating each one, repairs only structure a rule settles, and records what it fixed, disproved, and left for a person.
Agent Claude Code
Creates new spec areas, missing README files, and scaffolds Gherkin feature structure at explicitly specified paths under specs/. Use when adding a new app or library to the specs directory.
Agent Claude Code
Judges a running request-based interface with real requests under one charter, against its contract and behaviour specifications, and records criticality-rated findings with reproducing requests without fixing anything.
Agent Claude Code
Designs module boundaries, dependencies, and tradeoffs before implementation, judges finished work against that design, and serves as the architecture lens of a review pass, writing nothing but decision records.
Agent Claude Code
Resolves failing type checks, lint findings, and tests one cause at a time, diagnosing each before changing code, keeping pinned behaviour intact, and never suppressing or bypassing a check.
Agent Claude Code
Builds new or changed behaviour in named projects test-first under the adopted language-neutral and stack standards, and applies re-validated findings from a review, a test, or a frozen quality-gate ledger, never judging its own work.
Agent Claude Code
Decomposes a deterministic software goal into tasks, dispatches them to the software-engineering family in dependency order and in parallel, and merges the results until every goal check passes, editing nothing itself.
Agent Claude Code
Cuts versioned releases, deploys built artifacts to named environments, and repins pinned tools, each only through the repository's documented workflow, verifying the published result before reporting done.
Agent Claude Code
Audits code, interface component source, and scenario bindings in named projects against the adopted standards, including test-first evidence and regression tests, and returns rated findings without modifying anything.
Agent Claude Code
Evaluates a running web interface or command-line tool as a first-time user would, without its specifications, source, or designs, judging frozen tasks by named usability principles and recording severity-rated findings.
Agent Claude Code
Judges a running web interface in a real browser under one charter, against its specification, its approved design, or frozen exploratory charters, and records criticality-rated findings with reproduction steps without fixing anything.
Agent Claude Code
Audits one annotated-concept tutorial's mode, worked examples, and diagrams against the annotated-concept tutorial rules, with every product threshold read from the adopting repository's adapter, and returns criticality-rated findings without modifying anything.
Agent Claude Code
Executes Tutorial Annotated Concept Propagation on a frozen tutorial-annotated-concept ledger, repairing only what each worked example itself settles, with every threshold read from the adopting repository's adapter, and leaving authoring and pedagogy to the author.
Agent Claude Code
Audits one By Example tutorial's examples, annotations, and progression against the By Example tutorial rules, with every product threshold read from the adopting repository's adapter, and returns criticality-rated findings without modifying anything.
Agent Claude Code
Executes Tutorial By Example Propagation on a frozen tutorial-by-example ledger, repairing only what each example itself settles, with every threshold read from the adopting repository's adapter, and leaving authoring and pedagogy to the author.
Agent Claude Code
Audits one in-the-field guide set's scenario, built-in-first order, and production code against the in-the-field guide rules, with every product threshold read from the adopting repository's adapter, and returns criticality-rated findings without modifying anything.
Agent Claude Code
Executes Tutorial In the Field Propagation on a frozen tutorial-in-the-field ledger, repairing only what each guide itself settles, with every threshold read from the adopting repository's adapter, and leaving authoring and pedagogy to the author.
Agent Claude Code
Audits one primer's stated scope, capstone, and examples against the primer rules, with every product threshold read from the adopting repository's adapter, and returns criticality-rated findings without modifying anything.
Agent Claude Code
Executes Tutorial Primer Propagation on a frozen tutorial-primer ledger, repairing only what each example itself settles, with every threshold read from the adopting repository's adapter, and leaving authoring and pedagogy to the author.
Agent Claude Code
Researches current, verifiable information from the web in an isolated context. Use when you need facts beyond training data cutoff, latest API or library docs, current best practices, or verification of uncertain claims. Returns cited, structured findings without bloating main conversation context.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: