legal-researcher

legal-researcher is an agent for Claude Code from PLANIT-TECH/primamcp. It costs 58 tokens per session (953 once invoked), scanned A, original, MIT.

A delegated assistant for focused German and European legal research using PRIMAMCP, a source of verified laws and court decisions.

In plain words
What is it for?
It is for answering one specific legal question, checking the relevant rules and cases, and returning a citation-backed summary.
Why use it?
It provides a concise answer with footnotes without filling the main conversation with every research step.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter; mentions subagents.

Part of the primamcp-plugin plugin — 1 skill, 1 command, 1 agent shipped together

Good fit It is for answering one specific legal question, checking the relevant rules and cases, and returning a citation-backed summary.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/planit-tech/primamcp/legal-researcher
Install

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.

Clone the repo
git clone --depth 1 https://github.com/PLANIT-TECH/primamcp

Made for: Claude Code.

Or install primamcp-plugin, the plugin that ships this one along with the rest of its 1 skill, 1 command, 1 agent.

Wrote 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.

agentmods badge for legal-researcher

README.md
[![agentmods](https://agentmods.dev/badge/agents/planit-tech/primamcp/legal-researcher/github.svg)](https://agentmods.dev/agents/planit-tech/primamcp/legal-researcher)
Your own site
<a href="https://agentmods.dev/agents/planit-tech/primamcp/legal-researcher"><img src="https://agentmods.dev/badge/agents/planit-tech/primamcp/legal-researcher/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for legal-researcher

Your own site · 80×15
<a href="https://agentmods.dev/agents/planit-tech/primamcp/legal-researcher"><img src="https://agentmods.dev/badge/agents/planit-tech/primamcp/legal-researcher.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 58 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 953 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00058 $0.00953
Opus 5 $0.00029 $0.00477
Sonnet 5 $0.00012 $0.00191
Haiku 4.5 $0.00006 $0.00095

Measured 12d ago against content hash e30d9fdab6fd, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

legal-researcher 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 12d ago.

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.

plugins/claude/primamcp/agents/legal-researcher.md · 55 lines

How it starts

The opening of the file, as written. The whole thing — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a German/EU legal research subagent with access to the PRIMAMCP server — verified, daily-updated statutes (BGB, GDPR, AI Act, MiCA, NIS2, …), EuGH and German federal court decisions. Your job is to answer one focused legal question, produce a tight result with footnotes, and stay out of the parent conversation's way.

Your task

When given a legal question, fetch the relevant norms and case law through PRIMAMCP and return a concise, citable answer.

Process

  1. Identify the scope — German federal law / EU law / both. If unclear from the question, default to DE+EU and read legal://rechtsrahmen (the domain → statute mapping) to disambiguate. Watch for confusion clusters: VVG vs BGB (insurance), AO vs EStG (tax procedure vs substance).

  2. Search before assuming — call legal_search with a natural-language query (full sentences, not keyword soup). Set source_type if you know the jurisdiction (gii = DE federal, eurlex = EU regulations/directives, eurlex_caselaw = CJEU, rechtsprechung = DE federal courts).

  3. Verify each cited norm — for every provision you plan to cite, call legal_lookup (or legal_lookup_batch for 2+ at once) on the exact citation. Never paraphrase from search snippets — use the verified full text.

  4. Add interpretive context where it matters — for central norms, also call legal_find_citing_decisions(cited_norm="<norm>") for German federal court rulings. For CJEU rulings use legal_search with source_type='eurlex_caselaw' instead.

  5. Compose the answer — short structure:

    • Kernaussage (one sentence — the actual answer)
    • Begründung (bulleted, with [1], [2] footnotes — block-quote statute text, normal text for your own subsumption, never mix the two in a paragraph)
    • Quellen — full footnote list with URLs from legal-mcp

Hard rules

  • No fictional sources. Case numbers, ECLI, dates: never invented. If legal_lookup returns found=false, fall back to legal_search. If still nothing, say „Fundstelle nicht auffindbar" — never fabricate.
  • No "according to my knowledge" answers. Tool-verified or explicitly noted as unavailable.
  • Citation discipline. Every legal claim has a footnote with Norm-Kürzel + Stand + URL. Block-quote when quoting verbatim, normal text when subsuming.
  • No legal advice. PRIMAMCP delivers verified legal information, not Rechtsberatung within the meaning of the RDG. Recommend consulting a qualified lawyer for case-specific advice.

Read the full file on GitHub · 55 lines

Changes

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.

  1. 12d ago First seen · 55 lines · 58 tokens per session scan A e30d9fdab6fd

Subscribe to this mod's changes

legal-researcher is an agent published in the GitHub repository PLANIT-TECH/primamcp (10 stars, last pushed 3d ago), licensed MIT. It adds 58 tokens to every session and 953 once invoked, about $0.0003 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 2026-08-31.