Review contract
Clause-by-clause contract review — inventories key clauses (term, termination, liability cap, indemnity, IP, payment, confidentiality, governing law, auto-renewal), flags deviations from market-standard positions with RED/YELLOW/GREEN severity and verbatim quotes, drafts redline language, and sets negotiation priority. Use when the user says 'review this contract', 'what's wrong with this MSA', 'can we sign this as-is', or pastes any agreement for analysis.From its SKILL.md
npx -y skills add alebgl77/claude-inc --skill review-contractAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 8 stars8 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
5.2 KB, ~1.2k tokens by cl100k_base, as published. Nobody here has run it
Review Contract — Contract Reviewer
"Review any contract"
Works on pasted contract text or a file. No playbook required: deviations are measured against explicit market-standard positions, so every judgment is visible and overridable.
When to use
- "Review this MSA before tomorrow's call" — any contract that deserves more than a skim
- "Can we sign this vendor agreement as-is?" — a go/no-go read with fixes attached
- "What would you push back on here?" — building a negotiation position from scratch
- "They sent their redline back — does it hurt?" — re-review of a revised draft
- Standalone NDAs move faster through
triage-nda; use this skill when confidentiality terms sit inside a bigger deal
Workflow
- Ingest the contract (pasted text or file path). Record parties, which side we are, contract type, effective date, and any deadline mentioned.
- Inventory the nine key clauses: term, termination, liability cap, indemnity, IP ownership/license, payment, confidentiality, governing law and disputes, auto-renewal. A missing clause is a finding, not a blank — mark it MISSING.
- Benchmark each clause against market-standard positions, stated explicitly (e.g., liability capped at 12 months' fees and mutual; indemnity limited to third-party IP, bodily injury, and confidentiality claims; termination for cause with 30-day cure; auto-renewal with a 30-day-or-longer opt-out; no unilateral mid-term price increases).
- Assign severity per finding — RED: dealbreaker or uncapped exposure, do not sign as-is; YELLOW: off-market but fixable with targeted language; GREEN: market or better. Quote the operative text verbatim under every RED and YELLOW.
- Draft a redline for every RED and YELLOW: replacement language ready to paste into the document, not a description of what to change.
- Order negotiation priorities: REDs first, ranked by exposure; then YELLOWs ranked by cost-to-fix, marking which are trade material ("give to get").
- Assemble the report in the output format. When working in a folder, save it beside the contract as
<contract-name>-review.mdand report the path.
Output format
CONTRACT REVIEW — <contract name> — <date>
Parties: <counterparty> / <us> (we are the <customer/vendor/licensor/...>)
Type: <MSA / SaaS / services / license> Deadline: <date or none>
CLAUSE INVENTORY
| Clause | Present? | Severity | vs. market standard |
|-----------------|----------|----------|----------------------|
| Term | §<n> | GREEN | <one line> |
| Termination | §<n> | YELLOW | <one line> |
| Liability cap | MISSING | RED | uncapped by silence |
| Indemnity | §<n> | <sev> | <one line> |
| IP | §<n> | <sev> | <one line> |
| Payment | §<n> | <sev> | <one line> |
| Confidentiality | §<n> | <sev> | <one line> |
| Governing law | §<n> | <sev> | <one line> |
| Auto-renewal | §<n> | <sev> | <one line> |
FINDINGS
[RED] <clause> — <business impact, one line>
Quote: "<verbatim clause text>"
Redline: "<paste-ready replacement language>"
[YELLOW] <clause> — <business impact, one line>
Quote: "<verbatim clause text>"
Redline: "<paste-ready replacement language>"
[GREEN] <clauses at market or better — one line, no action>
NEGOTIATION PRIORITY
1. <RED finding> — must fix; walk away if refused
2. <RED finding> — must fix; fallback: <fallback position>
3. <YELLOW finding> — push; trade against <concession we can give>
*Issue-spotting support, not legal advice — engage counsel for binding decisions.*
Quality bar
- All nine key clauses inventoried; MISSING treated as a finding with severity
- Every RED and YELLOW quotes the clause verbatim — no paraphrase-only flags
- Every RED and YELLOW ships a paste-ready redline, not "negotiate this"
- Each severity justified in one line of plain business impact
- Negotiation order ranks by exposure and leverage, not page order
- Market-standard baseline stated wherever a deviation is flagged
Example
Invocation: "Review this SaaS agreement — we're the customer, they want it signed Friday." (contract pasted)
Produces: A review flagging RED on one-way indemnity ("Customer shall indemnify Provider against any and all claims arising from use of the Services") with a mutual, capped redline; YELLOW on a 12-month auto-renewal with a 90-day opt-out window; GREEN on payment and confidentiality. Negotiation priority puts indemnity first with a fallback cap at 12 months' fees, and marks the opt-out window as trade material.
Issue-spotting support, not legal advice — engage counsel for binding decisions.
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.
Gives 0 of the 12 instructions most review quality skills give in ~1.2k tokens
Counted across 1,048 of the 1,783 authors here whose files we hold, read 2026-08-07
- Ask questions one at a timein 81 of 1048, across 64 files
- Provide a recommended answer for each questionin 73 of 1048, across 50 files
- Explore the codebase instead of asking answerable questionsin 66 of 1048, across 42 files
- Resolve dependencies between decisions one-by-onein 42 of 1048, across 17 files
- Interview the user relentlessly about the planin 38 of 1048, across 13 files
- Order findings by severityin 31 of 1048
- Resolve each branch of the decision treein 27 of 1048, across 5 files
- Run a grilling sessionin 26 of 1048, across 5 files
- Update CONTEXT.md immediately when a term is resolvedin 26 of 1048, across 11 files
- Propose precise canonical terms for vague languagein 25 of 1048, across 7 files
- Create documentation files lazilyin 24 of 1048, across 5 files
- Assign severity to every findingin 24 of 1048
Said here and by no other author read
- inventory the nine key clauses
- mark missing clauses as findings
- benchmark clauses against market-standard positions
- assign RED YELLOW GREEN severity per finding
- quote operative text verbatim for RED and YELLOW
- draft paste-ready redline language
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.