Méthodologie « verification-first » pour la Cour Pénale Spéciale de Centrafrique (cour hybride, siège à Bangui). Toute citation est vérifiée (JusticeInfo, FIDH, HRW, legal-tools.org ; les sites officiels renvoient souvent une erreur 403). Couvre l'affaire Paoua, le « notamment » de l'article 3, et la complémentarité avec la CPI. Aide à la recherche, pas un conseil juridique. Fait partie de la bibliothèque open source « Skills for International Justice » — méthodologie : github.com/jeannesulzer/international-criminal-tribunals-skills
npx skills add https://github.com/lawve-ai/awesome-legal-skills --skill cps-rca-central-african-republic-jeanne-sulzer
This skill governs every output that touches the Cour Pénale Spéciale de la République Centrafricaine (CPS) — the Special Criminal Court in the Central African Republic. The discipline is simple and the reason for it is concrete: the CPS is the first fully hybrid court integrated into a national judicial system with jurisdiction over international crimes, and one of the only such courts operating in a country experiencing ongoing armed conflict. Its case numbers and procedural records are still developing; the Court conducts its principal work in French; and its complementarity relationship with the International Criminal Court (which has two situations in the CAR) is institutionally distinctive.
For any case-specific document — judgment, decision, indictment, filing, ordonnance d'instruction, requisitoire — verify before citing. "Verify" means web_fetch (or equivalent retrieval) to cpsrca.cf (the official CPS website) in the current conversation. Foundational texts in project knowledge (Loi organique 15.003, Loi 18.010 on the RPE, Code pénal centrafricain, Loi de coopération avec la CPI) are the exception; they may be cited directly. Nothing else.
In practice, cpsrca.cf is actively maintained (the CPS is operating with regular new audiences, decisions and press releases), but direct web_fetch to it — like several CPS-related sites (JusticeInfo, HRW) — often returns a 403 or partial content. Treat that as expected and work the fallback ladder (legal-tools.org, MINUSCA, JusticeInfo, RJDH, Radio Ndeke Luka) rather than as a dead end. Three levels:
Label the level where relevant. The CPS publishes summaries of judgments in addition to the full reasoned judgments; for substantive holdings, the arrêt motivé (full reasoned judgment) is authoritative, not the press release summary.
Step 0 — Identify the document. Before anything else, distinguish:
Step 1 — Plan citations. List every citation that will appear and the proposition each supports.
Step 2 — Verify with the fallback ladder. cpsrca.cf → JusticeInfo.net (Tier 2, for case summaries) → MINUSCA press releases (Tier 1 for institutional cooperation matters) → FIDH / Human Rights Watch monitoring reports (Tier 2) → ask the user.
Step 3 — Draft using verified material. Use the citation format in references/citation-format.md.
Step 4 — Self-audit. Each citation must trace to project knowledge or to a successful retrieval in this conversation.
references/foundational-texts.md on the debate over whether the list is illustrative rather than exhaustive.If not in project knowledge, retrieve from cpsrca.cf/documentations/textes-juridiques.
Tier 1 (authoritative):
minusca.unmissions.org) — institutional cooperation; some judgment announcementsTier 2 (secondary, must be labelled):
Never authoritative: Wikipedia, Grokipedia, social media, AI summaries.
CPS citations differ from international tribunal citations. Cases are referred to by name of the affair (not by case number in the international-tribunal sense) and chamber. The citation format follows French/Central African judicial conventions:
General form:
> *Procureur spécial contre [Defendant(s)]*, [name of the affair], Chambre [chamber], [Document type], [Date].
Worked examples:
See references/citation-format.md for the convention.
In either mode, Step 0 (identify the document and the affair) comes first. CPS proceedings move quickly between chambers (Chambre d'instruction → Chambre d'assises → Chambre d'appel) and the same defendants appear in multiple decisions across the proceedings — Step 0 prevents conflation.
The skill does not encode doctrine line by line. Starting points (each verified through the workflow before citing):
For each, verify through the workflow.
The CAR conflict (2003 to the present, with successive waves) involves Christian/animist (anti-balaka) and Muslim (Seleka, ex-Seleka, 3R, MPC, FPRC, UPC) armed groups, with extreme violence against civilians, sexual violence on a massive scale, displacement of about a fifth of the population, and ongoing fragility. The accused before the CPS are often members of armed groups still active or recently active. Witness protection is a major operational challenge. Victims and survivors are frequently in close proximity to perpetrators. Maintain factual precision, avoid sensationalism, use the language the Court itself uses (typically "groupe armé" rather than ethnic or religious shorthand), and never minimise. The Bangui Forum (2015) was a Central African national reconciliation process — referencing it as the political-institutional foundation of the CPS is appropriate.
references/authoritative-sources.md — source hierarchy and URLs (cpsrca.cf, JusticeInfo, MINUSCA, FIDH/OCDH)references/citation-format.md — citation conventions, named cases ("Affaire Paoua / Lemouna et Koundjili", "Affaire Ndélé 1", "Affaire Ndélé 1 contumace"), chamber designationsreferences/verification-workflow.md — fallback ladder, CPS-specific traps (national court vs international tribunal; CPS vs ICC; French as principal language; civil parties vs ICC victim participation; Affaire Paoua trial vs appeal sentence discrepancy; trial in absentia retrial right)references/foundational-texts.md — Loi organique 15.003, RPP (Loi 18.010), Code pénal centrafricain, MINUSCA cooperation MoUreferences/jurisprudence-map.md — topic-by-topic map of CPS holdings (small but growing as of 2026)examples/example-verification.md — verifying a Paoua case citation end-to-endexamples/example-audit.md — auditing user-supplied documents (Paoua trial vs appeal sentence; CPS misnomers)Assists in writing high-quality content by conducting research, adding citations, improving hooks, iterating on outlines, and providing real-time feedback on each section. Transforms your writing process from solo effort to collaborative partnership.
Identifies high-quality leads for your product or service by analyzing your business, searching for target companies, and providing actionable contact strategies. Perfect for sales, business development, and marketing professionals.
Use this skill to query your Google NotebookLM notebooks directly from Claude Code for source-grounded, citation-backed answers from Gemini. Browser automation, library management, persistent auth. Drastically reduced hallucinations through document-only responses.
Efficient database search tool for bioRxiv preprint server. Use this skill when searching for life sciences preprints by keywords, authors, date ranges, or categories, retrieving paper metadata, downloading PDFs, or conducting literature reviews.
Query and analyze scholarly literature using the OpenAlex database. This skill should be used when searching for academic papers, analyzing research trends, finding works by authors or institutions, tracking citations, discovering open access publications, or conducting bibliometric analysis across 240M+ scholarly works. Use for literature searches, research output analysis, citation analysis, and academic database queries.
Access USPTO APIs for patent/trademark searches, examination history (PEDS), assignments, citations, office actions, TSDR, for IP analysis and prior art searches.
Multiagent AI system for scientific research assistance that automates research workflows from data analysis to publication. This skill should be used when generating research ideas from datasets, developing research methodologies, executing computational experiments, performing literature searches, or generating publication-ready papers in LaTeX format. Supports end-to-end research pipelines with customizable agent orchestration.
Automated LLM-driven hypothesis generation and testing on tabular datasets. Use when you want to systematically explore hypotheses about patterns in empirical data (e.g., deception detection, content analysis). Combines literature insights with data-driven hypothesis testing. For manual hypothesis formulation use hypothesis-generation; for creative ideation use scientific-brainstorming.
Take lawve-ai/cps-rca-central-african-republic-jeanne-sulzer from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.