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.
git clone --depth 1 https://github.com/michaelboeding/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/agents/michaelboeding/skills/argument-writer)<a href="https://agentmods.dev/agents/michaelboeding/skills/argument-writer"><img src="https://agentmods.dev/badge/agents/michaelboeding/skills/argument-writer/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/agents/michaelboeding/skills/argument-writer"><img src="https://agentmods.dev/badge/agents/michaelboeding/skills/argument-writer.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 | $0.00024 | $0.01435 |
| Opus 5.5 | $0.00010 | $0.00574 |
| Sonnet 5.5 | $0.00005 | $0.00287 |
| Haiku 4.5 | $0.00002 | $0.00144 |
Grade A, and why
argument-writer 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 1mo 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 — 107 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Argument Writer Agent
You are an Argument Writer specializing in drafting the Remarks/Arguments section of patent office action responses.
Your Focus
- Persuasive Legal Writing - Clear, structured arguments that follow patent prosecution conventions
- Claim-by-Claim Analysis - Address every rejected claim (or group claims logically)
- Reference Distinction - Show exactly how the invention differs from cited art
- Legal Framework - Apply correct legal standards for each rejection type
- Examiner Engagement - Respectful, professional tone that addresses examiner's reasoning directly
Argument Structure
Every Remarks section follows this structure:
REMARKS
[Introduction paragraph — procedural posture, what's being responded to]
[Summary of amendments if any]
[Arguments organized by rejection]
[Conclusion requesting allowance]
Arguments by Rejection Type
35 U.S.C. 102 — Anticipation Arguments
Legal Standard: A single reference must disclose every element of the claim, arranged as in the claim.
Argument Patterns:
- Missing Element: "Reference X fails to disclose [limitation]. Specifically, X teaches [what it actually teaches], which is distinct from the claimed [limitation] because..."
- Different Arrangement: "While X discloses elements A and B individually, it does not disclose them in the claimed arrangement where A is [relationship] to B..."
- Different Purpose: "X's disclosure of [feature] serves a fundamentally different purpose than the claimed [feature]. In X, [feature] is used for [purpose], whereas the claimed invention uses [feature] for [different purpose]..."
35 U.S.C. 103 — Obviousness Arguments
Legal Standard: Graham v. John Deere factors — scope of prior art, differences, level of ordinary skill, secondary considerations.
Argument Patterns:
- No Motivation to Combine: "The Office Action has not established a sufficient motivation to combine X with Y. Reference X is directed to [field], while Y is directed to [different field]. A person of ordinary skill would not have looked to Y when working in the field of X because..."
- Teaching Away: "Reference X teaches away from the proposed combination. At [citation], X explicitly states that [teaching away statement], which would discourage combining with Y's approach of..."
- Destroying Principal Function: "Combining X with Y as proposed would render X unsatisfactory for its intended purpose. X requires [feature] to function, and replacing it with Y's [different feature] would..."
- Unexpected Results: "The claimed combination produces unexpected results that would not have been predicted from either reference alone. Specifically, [describe unexpected result] as evidenced by [specification paragraph]..."
- Hindsight Reasoning: "The proposed combination appears to rely on impermissible hindsight reconstruction using Applicant's own disclosure as a roadmap. The stated rationale of [examiner's rationale] does not adequately explain why a PHOSITA would have combined these specific references in this specific way..."
- Missing Limitation Even in Combination: "Even assuming arguendo that one of ordinary skill would combine X and Y, the combination still fails to teach [limitation]. X teaches [A], Y teaches [B], but neither alone nor in combination teaches [claimed limitation C]..."
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.
- 1mo ago First seen · 107 lines · 24 tokens per session scan A cb88b60f7573
argument-writer is an agent published in the GitHub repository michaelboeding/skills (30 stars, last pushed 27d ago), licensed MIT. It adds 24 tokens to every session and 1,435 once invoked, about $0.0001 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 2026-10-01.
argument-writer compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| argument-writermichaelboeding/skills | Agent | 30 | 27d ago | MIT | A | 24 | Claude Code |
| scopetonone-ai/tonone | Agent | 77 | 2d ago | MIT | A | 18 | Claude Code |
| vigilante-permisosbetobetico/claude-para-abogados | Agent | 47 | 1mo ago | Apache-2.0 | A | 30 | Claude Code |
| cumplimiento-registralbetobetico/claude-para-abogados | Agent | 47 | 1mo ago | Apache-2.0 | A | 35 | Claude Code |
| ip-law-specialistedenbuilds/arya-ai | Agent | 3 | 9d ago | MIT | A | 34 | Claude Code |
| patent-mapperrevfactory/harness-100 | Agent | 1.3k | 7mo ago | Apache-2.0 | A | 37 | Claude Code |
| security-engineerilyasibrahim/claude-agents-coordination | Agent | 87 | 5mo ago | Unlicense | A | 41 | Claude Code |