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 zgbrenner/agentcounsel --skill legal-prose-polishgit clone --depth 1 https://github.com/zgbrenner/agentcounselWrote 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/zgbrenner/agentcounsel/legal-prose-polish)<a href="https://agentmods.dev/skills/zgbrenner/agentcounsel/legal-prose-polish"><img src="https://agentmods.dev/badge/skills/zgbrenner/agentcounsel/legal-prose-polish/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/zgbrenner/agentcounsel/legal-prose-polish"><img src="https://agentmods.dev/badge/skills/zgbrenner/agentcounsel/legal-prose-polish.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 | $0.00050 | $0.01506 |
| Opus 5 | $0.00025 | $0.00753 |
| Sonnet 5 | $0.00010 | $0.00301 |
| Haiku 4.5 | $0.00005 | $0.00151 |
Grade A, and why
Legal Prose Polish 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 9d 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 — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Legal Prose Polish
Purpose
Polish draft legal work product so it reads like careful legal writing rather than generic AI output. The skill removes throat-clearing, filler transitions, fake balance, vague risk language, overlong caveats, unsupported certainty, conclusory statements, and awkward passive voice where responsibility matters.
This is a writing-quality pass, not a legal accuracy check. It must preserve legal meaning, citations, quotations, defined terms, privilege legends, client-specific instructions, formal legal tone where needed, cautious phrasing where substantively appropriate, and attorney-review warnings.
Use When
- A draft memo, email, demand letter, risk table, contract review, checklist, or client update sounds generic, padded, or AI-written.
- A draft overuses phrases such as "notably," "moreover," "it is important to note," "delve," "landscape," "underscore," "robust," or similar generic phrasing.
- The user asks to make a legal draft clearer, tighter, more lawyerly, or less AI-sounding.
- A draft contains conclusory or vague legal risk language that should be tied to facts, sources, or attorney-review caveats.
- A draft is external-facing and needs clearer agency, responsibility, and tone before attorney review.
Required Inputs
- The complete draft to revise.
- The intended audience and use: internal attorney, client, court, opposing party, regulator, business stakeholder, or other recipient.
- Any required style constraints, client instructions, defined terms, privilege designation, and non-editable language.
- Optional: source documents or authority list, if the polish pass must avoid altering cited propositions.
If the draft is not provided, stop and request it. Do not polish from a summary alone.
Do Not Use When
- The draft has not yet been substantively reviewed for source support, citations, assumptions, and attorney-review readiness.
- The user asks for legal advice, legal sign-off, or a conclusion that a position is correct.
- The task is to verify citations or factual claims; use
skills/legal-methodology/citation-integrity-check/SKILL.mdorskills/legal-methodology/source-validation/SKILL.md. - The task is to change legal substance without attorney-provided direction.
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.
- 9d ago First seen · 105 lines · 50 tokens per session scan A 54411e6fe5b7
Legal Prose Polish is a skill published in the GitHub repository zgbrenner/agentcounsel (19 stars, last pushed 1mo ago), licensed MIT. It adds 50 tokens to every session and 1,506 once invoked, about $0.0003 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 2026-09-03.
Other skills, from other repositories
legal-research
A structured system for legal research involving laws, court decisions, academic sources, arguments, and evidence. It separates facts, legal claims, scholarly views, and the author's own analysis.
legal-research-authorship
A Chinese-language review skill for legal papers that checks formulaic wording, unsupported claims, source use, and the author's reasoning. It is designed to improve clarity without pretending to prove who wrote the text or bypass AI detectors.
legal-research-providers
A connector for authorised legal databases, including databases such as Pkulaw that provide laws, court decisions, and academic literature. It imports results from MCP or command-line tools into a structured research record.
legal-research-claims
A legal-argument mapping guide that breaks a law question or paper claim into testable statements, supporting evidence, opposing evidence, and limits.
legal-research-empirical
A Chinese-language skill for designing empirical studies of court decisions and legal texts. It can import DOCX, TXT, or Markdown documents, define coding rules, track samples, and support reproducible analysis.
legal-research-sources
A legal research workflow for finding and verifying laws, regulations, court decisions, academic papers, and official data.