legal-hold

legal-hold is a skill for Claude Code from anthropics/claude-for-legal. It costs 68 tokens per session (3,512 once invoked), scanned A, original, Apache-2.0.

A legal-hold workflow for telling people to preserve information that may be relevant to a legal dispute. It can issue, refresh, release, and report on holds while tracking their status.

In plain words
What is it for?
Drafting hold and release notices, recording custodians and date ranges, scheduling refreshes, and reporting hold status across matters.
Why use it?
It helps prevent relevant records from being deleted and keeps hold instructions current as a matter changes.

Skill for Claude Code ✓ vendor

Written for Claude Code: argument-hint in frontmatter. Also seen: reads .claude/ paths; mentions CLAUDE.md.

Part of the litigation-legal plugin — 16 skills, 1 agent shipped together

Good fit Drafting hold and release notices, recording custodians and date ranges, scheduling refreshes, and reporting hold status across matters.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/anthropics/claude-for-legal/legal-hold
About the project

Claude for Legal is a collection of AI agents, skills, and data connectors for legal work such as commercial, privacy, employment, litigation, regulatory, and intellectual-property matters. Lawyers and legal teams use it through Claude Cowork, Claude Code, or the Claude Managed Agents API, with attorney review required for every output. The catalogue entries are its workflow plugins, skills, agents, and connectors.

anthropics/claude-for-legal · 9,529 stars · on GitHub

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.

Any agent
npx skills add anthropics/claude-for-legal --skill legal-hold
Clone the repo
git clone --depth 1 https://github.com/anthropics/claude-for-legal

Made for: Claude Code.

Or install litigation-legal, the plugin that ships this one along with the rest of its 16 skills, 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-hold

README.md
[![agentmods](https://agentmods.dev/badge/skills/anthropics/claude-for-legal/legal-hold/github.svg)](https://agentmods.dev/skills/anthropics/claude-for-legal/legal-hold)
Your own site
<a href="https://agentmods.dev/skills/anthropics/claude-for-legal/legal-hold"><img src="https://agentmods.dev/badge/skills/anthropics/claude-for-legal/legal-hold/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-hold

Your own site · 80×15
<a href="https://agentmods.dev/skills/anthropics/claude-for-legal/legal-hold"><img src="https://agentmods.dev/badge/skills/anthropics/claude-for-legal/legal-hold.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 68 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,512 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. Third-party audits
  • Socket pass 14 Sept 2026
  • Snyk pass 14 Sept 2026
How audits are shown
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 $0.00068 $0.03512
Opus 5.5 $0.00027 $0.01405
Sonnet 5 $0.00014 $0.00702
Haiku 4.5 $0.00007 $0.00351

Measured 3d ago against content hash 42fba1356045, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-26, from the pricing page.

Security

Grade A, and why

legal-hold 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 3d 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.

litigation-legal/skills/legal-hold/SKILL.md · 239 lines

How it starts

The opening of the file, as written. The whole thing — 239 lines — stays where its author put it; the contents beside it link to each section on GitHub.

  1. If --status (no slug): read _log.yaml, produce portfolio-wide hold report.
  2. Otherwise: load ~/.claude/plugins/config/claude-for-legal/litigation-legal/matters/[slug]/matter.md + log row.
  3. Load ~/.claude/plugins/config/claude-for-legal/litigation-legal/CLAUDE.md → privilege markings, hold template pointer, escalation norms.
  4. Follow the workflow and reference below.
  5. Route by flag:
    • --issue: capture scope, custodians, date range, systems. Draft legal-hold-v1.docx. Update legal_hold fields. Append history entry. Set next_refresh (default +6mo).
    • --refresh: capture scope/custodian changes. Draft next version. Update last_refresh + next_refresh. Flag departed custodians.
    • --release: capture release date, retention instruction. Draft release notice. Set released: field.
  6. Confirm before writing. Show the user the draft notice and the log diff.

Purpose

A legal hold is the most mechanical high-stakes document in-house counsel writes. The notice itself is templated. The failure modes are operational: issued too late, scoped too narrowly, never refreshed, never released. This skill owns all four phases: issue → refresh → (release) → track.

The portfolio already flags missing holds; this skill writes them.

Jurisdiction assumption

Preservation duties vary materially by forum. Federal common law (via Zubulake / Residential Funding / Rule 37(e)) differs from state practice; states differ from each other on trigger timing, scope, sanctions, and spoliation remedies; regulatory preservation obligations overlay civil rules in some matters (SEC Rule 17a-4, HIPAA, etc.). The trigger, scope, and sanctions exposure cited in the draft are a starting-point read for the forum named in the matter — confirm with counsel before issuing, refreshing, or releasing.

Load context

  • ~/.claude/plugins/config/claude-for-legal/litigation-legal/matters/_log.yaml — log row (legal_hold fields + status)
  • ~/.claude/plugins/config/claude-for-legal/litigation-legal/matters/[slug]/matter.md — matter context (counterparty, facts, key custodians from internal_owners)
  • ~/.claude/plugins/config/claude-for-legal/litigation-legal/CLAUDE.md — house style for litigation hold template pointer, privilege marking, escalation norms

Read the full file on GitHub · 239 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. 3d ago First seen · 239 lines · 68 tokens per session scan A 42fba1356045

Subscribe to this mod's changes

legal-hold is a skill published in the GitHub repository anthropics/claude-for-legal (9,529 stars, last pushed yesterday), licensed Apache-2.0. It adds 68 tokens to every session and 3,512 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-09-23.

Related

Other skills, from other repositories

html-ppt-zhangzara-stencil-tablet

A workplace-safety compliance review for a manufacturing regulator — findings, the evidence chain, and the corrective mandate. Built as a decision-grade policy briefing deck for regulator, plant leadership.

nexu-io/open-design · 49 tokens

treatment-plans

Format and structurally validate local treatment-plan documentation after clinical decisions have already been supplied and verified by authorized licensed professionals. Use for source traceability, clinician-authored intervention records, goals and checkpoints, shared-decision records, reconciliation handoffs, and…

K-Dense-AI/scientific-agent-skills · 62 tokens

software-copyright-materials

A workflow for creating Chinese application materials for software copyright registration from a real software project.

Fokkyp/SoftwareCopyright-Skill · 98 tokens

china-kyc-doc-parse

A guide for extracting and checking information from Chinese identity documents used in customer verification. It covers personal identity cards, business licences, and related company or identity details.

jwangkun/claude-for-financial-services-cn · 88 tokens

iflytek-contract-intelligence-review

A contract-review workflow for scanned or digital agreements. It can recognize document text, examine clauses, detect risks, check compliance, and create translation summaries; its output is for review and is not formal legal advice.

iflytek/iFly-Skills · 116 tokens

legal-writer

Legal document drafting -- contracts, memos, briefs, complaints, demand letters, opinions, discovery, settlements, ToS, privacy policies. Full pipeline: document structure, per-section writing, Bluebook citation, case law lookup (CourtListener API), regulation lookup (eCFR API), DOCX output, and TDD-style verification…

kortix-ai/suna · 161 tokens