medelmouhajir/cndp-compliance-checklist

A curated compliance self-assessment assistant built on Morocco's Law 09-08 (Data Protection) and CNDP requirements, designed under the **ISLI Universal Skill Runtime (USR) v2.0 spec** as an isolated HTTP microservice. # CNDP Compliance Checklist Skill (ISLI AI v2.0) A curated compliance self-assessment assistant built on Morocco's Law 09-08 (Data Protection) and CNDP requirements, designed under the **ISLI Universal Skill Runtime (USR) v2.0 spec** as an isolated HTTP microservice. This skill reinforces local data sovereignty by enabling users and client agents to self-evaluate data protection compliance ahead of deployments. --- ## Features & Exposed Tools The skill registers three key tools with the ISLI Agent Core: 1. **`get_compliance_checklist`** - **Endpoint:** `POST /get-checklist` - **Description:** Returns the complete self-assessment question set grouped into five major categories (Consent, Formalities, Data Subject Rights, Security, and Cross-border Transfers). 2. **`evaluate_compliance`** - **Endpoint:** `POST /evaluate` - **Description:** Performs rules-based evaluation using processed data attributes (sensitive fields, hosting locations, and existing policies) to output: - The corresponding compliance risk rating (**Low**, **Medium**, or **High**). - Required formalities (e.g. Standard prior declaration vs. strict prior authorization requests). - Customized action recommendations. 3. **`search_cndp_rules`** - **Endpoint:** `POST /search-rules` - **Description:** Instantly queries the static knowledge base of key Law 09-08 Articles (Articles 4, 5, 7, 8, 9, 12, 21, 22, 23, 43, 44) by keywords. --- ## Technical Architecture The codebase contains: - `isli-skill.yaml` - The skill manifest mapping runtime targets and JSON Schema parameters. - `main.py` - FastAPI-based microservice embedding the legal compliance framework, token validation, and the REST endpoints. - `Dockerfile` - Builds the lightweight Python environment for containerized execution. ### Local Development 1. Install requirements: ```bash pip install -r requirements.txt ``` 2. Start the Uvicorn dev server: ```bash uvicorn main:app --reload --port 8000 ``` 3. Test locally using curl or swagger do …