Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add jonashertner/opencaselaw --skill citation-checkgit clone --depth 1 https://github.com/jonashertner/opencaselawWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/jonashertner/opencaselaw/citation-check)<a href="https://agentmods.dev/skills/jonashertner/opencaselaw/citation-check"><img src="https://agentmods.dev/badge/skills/jonashertner/opencaselaw/citation-check.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00046 | $0.00920 |
| Opus 5 | $0.00023 | $0.00460 |
| Sonnet 5 | $0.00009 | $0.00184 |
| Haiku 4.5 | $0.00005 | $0.00092 |
Grade A, and why
citation-check scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Citation check with ocl
Use this whenever a draft, memo, brief or answer cites Swiss court decisions (BGE/ATF/DTF, BGer dockets such as 4A_747/2012, cantonal dockets) or quotes an Erwägung. Existence is checked against the corpus; legal support is not.
Steps
- If the draft is a file (.docx, .md, .html, .txt), run
ocl check <draft> --format json --no-report > check.json: the citations and the quotations next to them are found in the prose and checked in one go (install:pipx install opencaselaw-clioruv tool install opencaselaw-cli). Otherwise extract every citation intoreferences.jsonl, one JSON line each, exactly as the author wrote it, long form included:{"reference": "BGer 4A_747/2012 vom 5. April 2013, E. 3.2"}, with aquotefield for quoted passages. - Run
ocl citations resolve --input references.jsonl --format jsonl > resolution.jsonlfor a list; the rows have the same shape ascheck'sresults.checkalso returnsstatutes(Art./§ references:statute_found,article_missing,article_empty,law_unknown,unverifiable,error) andunparsed(citation-like strings that were not checked); report both. - Read every row.
resolvedmeans the decision exists and carries a label the author wrote (and the pinpoint, if any, exists:pinpoint_status: retrieved).pinpoint_unavailablemeans the decision exists but the numbered passage is not indexed; withpinpoint_status: parent_retrievedthe parent Erwägung's text is in the row and the lettered part must be located inside it.discrepancymeans the decision exists but the date or the docket the author wrote next to it is wrong: reportdiscrepanciesverbatim.missingmeans no such decision in the corpus: flag it for the author; never replace it with a "close" citation.ambiguousmeans several decisions carry the label: name the court, ask which one, or use thedecision_id.unrecognizedmeans the service proposed a decision that does not carry the author's label: treat it as unverified, and never reportservice_candidateas the citation. - For quoted passages, run
ocl quotes check --input quotes.jsonl --format jsonlwith one row per quotation (reference,pinpointwhen cited,quoteexactly as the author wrote it).exactstands; fornearreport the listed differences and replace the quotation with the served wording; fornot_foundtell the author the quotation is not in the served text that was compared (the closest served text is shown); forunverifiable(no served text: offline without an indexed pinpoint) say the quotation was not checked and must be checked against the decision, never that it is not there. Do not paraphrase inside quotation marks. - Report per citation: reference, status, decision_id, canonical citation string from the service, pinpoint status, discrepancies. State that this establishes existence and wording only, not that the authority supports the proposition.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- today Changed · +10 lines fbab479c58e2
- 2d ago First seen · 51 lines · 46 tokens per session scan A 7d46e19c89e3
citation-check is a skill published in the GitHub repository jonashertner/opencaselaw (61 stars, last pushed yesterday), licensed MIT. It adds 46 tokens to every session and 920 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-06.
Other skills, from other repositories
legaltech-expert
Expert in legal technology: contract lifecycle management, e-discovery, AI-assisted legal research, case and matter management, document automation, and compliance tooling. Use when the user mentions legal case or matter management, e-discovery and legal holds, CLM or contract review and obligation tracking, legal…
legalize-kr
Use this skill when working with Legalize-KR public Korean legal data, including laws, court precedents, administrative rules, and local ordinances. It guides agents to choose between legalize-cli, the local stdio legalize MCP server, git clone workflows, and direct GitHub access for search, retrieval, article lookup…
format-monitoring-report
A formatting guide for turning already-approved legal-monitoring content into a DOCX or PDF report. It specifies a seven-section layout, blue labels, Times New Roman, footnotes, page numbers, and a final comparison table.
verifiable-legal-document
A Russian-language workflow for producing legal documents with claim identifiers, linked supporting passages, hashes, and validation reports.
matter-intake
A setup process for a new legal matter or research assignment. It turns an unstructured request into a defined question, scope, data classification, plan, sources, deadline, and completion criteria.
ukrainian-legal-research
A research workflow for determining which Ukrainian laws and regulations applied on a specific date. It checks official wording, versions, effective dates, transitional rules, legal hierarchy, authority, and official sources.