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 LegalQuants/lq-skills --skill vibe-legal-batch-redlinergit clone --depth 1 https://github.com/LegalQuants/lq-skillsWrote 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/legalquants/lq-skills/vibe-legal-batch-redliner)<a href="https://agentmods.dev/skills/legalquants/lq-skills/vibe-legal-batch-redliner"><img src="https://agentmods.dev/badge/skills/legalquants/lq-skills/vibe-legal-batch-redliner/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.
<a href="https://agentmods.dev/skills/legalquants/lq-skills/vibe-legal-batch-redliner"><img src="https://agentmods.dev/badge/skills/legalquants/lq-skills/vibe-legal-batch-redliner.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00043 | $0.02793 |
| Opus 5 | $0.00022 | $0.01396 |
| Sonnet 5 | $0.00009 | $0.00559 |
| Haiku 4.5 | $0.00004 | $0.00279 |
Grade A, and why
vibe-legal-batch-redliner 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.
How it starts
The opening of the file, as written. The whole thing — 198 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Vibe Legal Server — Batch Contract Redlining
CRITICAL — Data flow warning. This skill sends the full text of every uploaded contract to Google Gemini for AI-driven redlining. The "bring your own key" model means traffic is keyed to your account, but the document body still leaves your environment and is processed on Google infrastructure. Do not invoke on privileged client material unless your firm has approved sending legal documents to Google Gemini under the relevant DPA / firm AI policy. If your firm has not approved Gemini for client work, configure the server to point at an on-prem model (e.g., a local Llama / Mistral deployment) or an alternative provider covered by your firm's policy (e.g., Azure OpenAI under your tenant's data-processing terms) before running the skill on real matters.
Audience and Work Shape
Audience: UK-qualified commercial lawyers (or supervised paralegals under a qualified lawyer's sign-off) who are comfortable reading tracked changes line-by-line and willing to reject AI-applied edits. Not for non-lawyer end-users, self-service contract-acceptance workflows, or unsupervised paralegal review.
Work shape: Bounded Transactional at batch scale. Each redlined DOCX is a draft proposal, not a recommendation, and the operator must accept or reject every tracked change individually. The work shape note matters because batch volume (up to 5 documents in one run) creates exactly the pressure that produces skim-acceptance — at scale, escalation matters more, not less. If you cannot honestly commit to per-change review for every document in the batch, do not run the batch.
Scope and Legal Use
This skill provides legal support (a first-pass redline against a playbook), not legal advice. Three legal failure modes the operator must own:
- Advice vs. support. The redlines are positional drafts; the responsible lawyer owns every position taken in the output. A tracked change in the DOCX is not a recommendation by the skill, by LegalQuants, or by the original author.
- Privilege and confidentiality. Uploading a client contract to Google Gemini (via your BYOK API key) routes the document text through a third-party AI processor and, depending on Gemini configuration and region, may involve cross-border processing. This can impair privilege and breach engagement-letter / client-confidentiality obligations absent client consent and firm AI-policy coverage. Treat every invocation as an outbound disclosure and confirm coverage before upload.
- Accountability. A named reviewing lawyer must sign off on the final redline. The skill produces no audit trail of which AI operations were accepted or rejected; the reviewing lawyer is responsible for capturing that record outside the tool.
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.
- 12d ago First seen · 198 lines · 43 tokens per session scan A 432b2405f392
vibe-legal-batch-redliner is a skill published in the GitHub repository LegalQuants/lq-skills (54 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 43 tokens to every session and 2,793 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 2026-08-30.
Other skills, from other repositories
document-fill
A document-filling assistant that uses a supplied template and case files to fill in forms, contracts, or legal documents. It also reports where each filled value came from and which details are missing.
openreview-cli
Use when a user wants to review, compare, search, or analyze legal/contract documents (PDF/DOCX) locally — produce review memos, compare two versions, search indexed clauses, run negotiation analysis, manage LLM gateway slots/playbooks, or audit/delete stored PII mappings. This is the local openreview CLI, NOT the…
generate_fillable_contract_html
Chinese HTML templates for contracts, quotations, and authorization letters with blank fields for later completion or DOCX conversion.
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.
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…
handbook-updates
Diff a proposed handbook change against the current version, flag ripple effects and state supplement impacts. Use when user says "update the handbook", "add this to the handbook", "handbook change", or has a policy ready for insertion.