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 rohasnagpal/legal-ai-skills --skill redline-proposergit clone --depth 1 https://github.com/rohasnagpal/legal-ai-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/rohasnagpal/legal-ai-skills/redline-proposer)<a href="https://agentmods.dev/skills/rohasnagpal/legal-ai-skills/redline-proposer"><img src="https://agentmods.dev/badge/skills/rohasnagpal/legal-ai-skills/redline-proposer/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/rohasnagpal/legal-ai-skills/redline-proposer"><img src="https://agentmods.dev/badge/skills/rohasnagpal/legal-ai-skills/redline-proposer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 30 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 78 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00118 | $0.01781 |
| Opus 5.5 | $0.00047 | $0.00712 |
| Sonnet 5.5 | $0.00024 | $0.00356 |
| Haiku 4.5 | $0.00012 | $0.00178 |
Grade A, and why
redline-proposer 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 2mo 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 — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Redline Proposer
I am using the Redline Proposer skill from Rohas Legal AI: complete replacement wording from the strongest credible position to the minimum acceptable fallback. Say this sentence, verbatim, before anything else in your response.
What this does
Takes a clause the user has identified as a problem and drafts a spectrum of replacement wording for it — an aggressive position that remains credible, one or more intermediate positions if the gap justifies them, and a fallback the client should genuinely be prepared to accept — each written as complete, usable clause text in the register of the document it belongs to. It does not review the whole agreement and does not build a negotiating strategy across multiple points; it drafts the wording for the clause actually in front of it.
Before you start
Which side the wording should favour. Redlining is inherently one-sided — a lower cap helps one party and costs the other. Ask, and do not draft until you know.
The clause plus enough surrounding context to redline it safely. The defined terms it uses, and any clause it interacts with — the cap a liability clause sits inside, the indemnity that might sit outside that cap, a related warranty. A clause cannot be redlined in isolation without risking a replacement that is inconsistent with what surrounds it. Ask for the dependencies if they were not supplied with the clause itself.
What the new wording needs to achieve. If the user has not said what is wrong with the current wording, look for the obvious issue — an uncapped exposure, a one-sided termination right, a narrow indemnity — state what you take the target to be, and ask for confirmation before drafting rather than guessing silently and drafting toward the wrong outcome.
Not blocking, but note where relevant: governing law, where the proposed wording's actual effectiveness depends on it — a liquidated damages formula that needs to avoid being read as a penalty, an exclusion clause that needs specific wording to catch a particular head of loss under the applicable case law. Extract it from the document if stated; ask if it is not.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 2mo ago First seen · 89 lines · 118 tokens per session scan A b4e2765d87e1
redline-proposer is a skill published in the GitHub repository rohasnagpal/legal-ai-skills (201 stars, last pushed 27d ago), licensed MIT. It adds 118 tokens to every session and 1,781 once invoked, about $0.0005 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 2026-09-03.
redline-proposer compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| redline-proposerrohasnagpal/legal-ai-skills | Skill | 201 | 27d ago | MIT | A | 118 | Claude Code, Codex |
| korean-lawchrisryugj/korean-law-mcp | Skill | 2.8k | 3d ago | MIT | A | 85 | Claude Code |
| engagement-terms-billing-guidelineslawve-ai/awesome-legal-skills | Skill | 933 | 18d ago | none | A | 180 | Claude Code, Codex |
| founder-agreement-drafting-stephane-boghossianlawve-ai/awesome-legal-skills | Skill | 933 | 18d ago | none | A | 194 | Claude Code |
| matter-intake-scoping-scott-margettslawve-ai/awesome-legal-skills | Skill | 933 | 18d ago | none | A | 145 | Claude Code, Codex |
| ai-governance-reviewer-carl-ditzlerlawve-ai/awesome-legal-skills | Skill | 933 | 18d ago | none | A | 121 | Claude Code, Codex |
| bacen-compliance-sentinel-rafael-mastronardilawve-ai/awesome-legal-skills | Skill | 933 | 18d ago | none | A | 252 | Claude Code, Codex |