What does this skill do, and when should you use it?
This is one of 14 skills in the AI Legal Assistant for Claude Code collection, located at skills/legal-negotiate/SKILL.md. Triggered by /legal negotiate <file>, it reads a contract file, pasted text, or URL and scans every clause for terms unfavorable to the user, rating each HIGH/MEDIUM/LOW risk. For each issue it produces the original quote, a plain-English problem statement, complete replacement clause text, a persuasive talking point, and an acceptance-likelihood estimate, grouped into MUST-CHANGE / SHOULD-CHANGE / NICE-TO-CHANGE tiers. The final output is a negotiation-strategy Markdown report with a concession strategy and a professional email template, and it explicitly disclaims legal advice.
Reads local contract files via the Read tool, fetches URLs via WebFetch, or uses pasted text; identifies contract type and parties; flags one-sided clauses, excessive liability, vague language, missing protections, and traps like auto-renewal, assigning HIGH/MEDIUM/LOW RISK; for each issue generates precise replacement clause text (not vague suggestions), collaborative talking points, and acceptance likelihood (High/Medium/Low); tiers findings into MUST-CHANGE, SHOULD-CHANGE, and NICE-TO-CHANGE with a suggested concession order; drafts a professional, non-adversarial revision email; and writes a NEGOTIATION-STRATEGY-[contract-name].md file.
- A freelancer preparing a prioritized list of clause changes and talking points before signing a client contract
- A small business owner who received a vendor contract and needs replacement wording for liability caps and auto-renewal terms
- An agency drafting a polished, sendable revision email to the counterparty's legal department
- A negotiator close to signing who needs a pre-planned concession order: what to give up first, what never to concede
- A contract party facing vague indemnification language who needs an insert-ready clause, not a generic recommendation
How do you install this skill?
- Output is informational only and not legal advice; all replacement clauses must be reviewed by a licensed attorney before signing.
- URL fetching via WebFetch and contract processing send sensitive commercial text through the AI pipeline; use caution with confidential agreements.
- Skill name (counter-proposal-generator) mismatches the command (legal-negotiate); rely on the README command.
- No license or version metadata and unclear maintenance/update path; accuracy of clause identification and replacement language is unverified.
- English-only with undeclared jurisdiction; not suitable for Chinese-language contracts or non-US/UK legal contexts.
- Network access
- Local filesystem
Install the full collection (this skill ships with it): curl -fsSL https://raw.githubusercontent.com/zubair-trabzada/ai-legal-claude/main/install.sh | bash. This installs all 14 skills, 5 agents, and the PDF generation scripts. Standalone installation of this single skill is not documented; the license is unknown.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/zubair-trabzada/ai-legal-claude.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/skills/legal-negotiate" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
In Claude Code, run /legal negotiate <file> where <file> is a local file path, pasted contract text, or a URL. The skill writes NEGOTIATION-STRATEGY-[contract-name].md in the input file's directory (or the current working directory for pasted text).
What are this skill's strengths and limitations?
- Output is immediately usable: replacement clauses are written as complete, insert-ready text rather than vague advice like 'add a liability cap'
- Three priority tiers plus an explicit concession order mirror real negotiation workflow, including strategic bargaining chips
- Includes a professional, collaborative email template ready to send to the other party
- Each proposal carries an acceptance-likelihood estimate to help anticipate pushback
- Output includes a clear legal disclaimer, positioning the tool as informational rather than legal advice
- Not a lawyer and cannot replace review by a licensed attorney — the output itself repeats this repeatedly
- Triggered only via the /legal negotiate slash command, which depends on Claude Code's command mechanism; other platforms require rework of the trigger
- Acceptance-likelihood ratings are estimates with no stated calibration basis or test data
- Repository has no stated license, so commercial-use terms are unclear
- The body references Claude Code's Read and WebFetch tools; README metrics (60-second reviews, savings versus lawyer fees) are marketing claims without independent verification
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Counter-Proposal Generator this page | 41 · Not recommended | ★ 1.8k | 7mo ago | — |
| Missing Protections Finder (AI Legal Assistant) | 45 · Not recommended | ★ 1.8k | 7mo ago | — |
| AI Legal Assistant (legal/SKILL.md Orchestrator) | 43 · Not recommended | ★ 1.8k | 7mo ago | — |
| Contract Comparison Skill | 45 · Not recommended | ★ 1.8k | 7mo ago | — |
| Plain English Contract Translator | 44 · Not recommended | ★ 1.8k | 7mo ago | — |
How did FollowSkills review this skill?
Pure prompt-based skill with least-privilege scope (reads user-specified files, optional WebFetch, writes one markdown file); prominent legal disclaimer included; no covert exfiltration or dangerous defaults. Deducted for: URL fetching without confirmation, unexplained data flow for sensitive contract text, no rollback/overwrite handling, unverified publisher identity.
Instructions are self-consistent with clear steps; the happy path (file path or pasted text) is plausible. Deducted for: no tests or sample inputs/outputs, no error handling or failure feedback for unreachable URLs, missing files, empty or non-contract input; no evidence of deterministic clause identification quality.
Trigger is explicit (/legal negotiate <file>) and the scenario (pre-signing negotiation prep) is clear with a complete output structure. Deducted for: undeclared capability boundaries (jurisdiction, English-only assumption), no Chinese-language support, no statement of applicable contract types or non-fit ranges.
Documentation is well layered and templates are complete and readable. Deducted for: frontmatter name (counter-proposal-generator) inconsistent with directory (legal-negotiate) and README command; no versioning, changelog, known-limitation disclosure, FAQ, or license metadata.
Promises directly usable replacement clauses, talking points, and an email template with concession strategy; some marginal value over manual drafting. Deducted for: static review cannot verify legal correctness of proposed clause language; high cost of legal errors means output will require attorney review, so direct-usability evidence is limited.
Only author-written instructions and templates; no tests, sample outputs, third-party execution evidence, or user corroboration. Auditable primary material is thin under static review, so the score stays low.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →