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 isvlasov/rageatc-oss --skill communicating-in-writinggit clone --depth 1 https://github.com/isvlasov/rageatc-ossWrote 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/isvlasov/rageatc-oss/communicating-in-writing)<a href="https://agentmods.dev/skills/isvlasov/rageatc-oss/communicating-in-writing"><img src="https://agentmods.dev/badge/skills/isvlasov/rageatc-oss/communicating-in-writing/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/isvlasov/rageatc-oss/communicating-in-writing"><img src="https://agentmods.dev/badge/skills/isvlasov/rageatc-oss/communicating-in-writing.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 | $0.00055 | $0.01273 |
| Opus 5 | $0.00028 | $0.00636 |
| Sonnet 5 | $0.00011 | $0.00255 |
| Haiku 4.5 | $0.00006 | $0.00127 |
Grade A, and why
communicating-in-writing 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 11d 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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Communicating in Writing
Advisory and high-freedom: the layers below are thinking tools, not a sequence to execute. Work through intent, audience, and structure before drafting prose; craft comes last. Structure determines comprehension more than prose quality does — fix structure before fixing sentences.
Layer 1 — Intent
Establish before writing a word:
- Purpose — what the document must achieve, in one sentence. If there are multiple outcomes, rank them; a document with two equal purposes usually fails both.
- What changes — what the reader should know, believe, feel, or do after reading that they did not before. If you cannot state this, the writing is not ready to begin.
- Call to action — what you are asking the reader to do. Even informational documents have an implicit next step; surface it.
- Audience, document type, constraints — who reads it, what form it takes, what is non-negotiable.
Resolve all of these before drafting. A document begun without a clear purpose cannot be edited into shape — it needs rewriting once purpose is found.
Layer 2 — Audience
- Identify the primary reader (the decision-maker) and any secondary readers. For each: what do they know, what do they care about, what will they resist?
- Working memory holds roughly four new chunks at once. One idea per paragraph, lists of seven or fewer items, complex arguments broken into discrete steps.
- Prior knowledge determines structure, not just vocabulary. Experts tolerate dense sections and domain compression; non-experts need the same content in more sections, with scaffolding, analogy, and explicit transitions. Calibrate density to the reader, not to what feels natural to you as the expert.
- Mismatch signals: jargon the reader does not know, assumed context they do not have, detail level misaligned (too granular for an executive; too high-level for an implementer).
Layer 3 — Structure
- Lead with the answer. Conclusion or recommendation first, then reasoning and evidence. Exception — deliberate, not default: when the reader needs emotional or contextual preparation (bad news, controversial recommendations), build the case first.
- Each section answers exactly one question; split any section answering two. Headings signal what the section delivers, not its topic — "Three decisions required by Friday", not "Next steps".
- Test the through-line: can you state the document's argument in two or three sentences, in order? If not, the structure has a gap. For persuasive documents: situation → complication → question → answer.
- Persuasive documents work on three levels: sound argument (logos), credibility established early rather than assumed (ethos), connection to what the reader actually values (pathos). Most writing defaults to logic alone — check the other two explicitly.
- Readers remember openings and closings most. Open with the core message; close by reinforcing it or driving the call to action, never with a generic summary.
- For longer documents: summary up front, parallel structure in lists and headings, key terms repeated consistently — variation for style creates confusion, consistency aids navigation.
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.
- 11d ago First seen · 65 lines · 55 tokens per session scan A 9def571524f3
communicating-in-writing is a skill published in the GitHub repository isvlasov/rageatc-oss (9 stars, last pushed 1mo ago), licensed MIT. It adds 55 tokens to every session and 1,273 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 2026-08-31.
Other skills, from other repositories
design-method
Use when starting any design task — resolves register, sets the 3 dials, locks Gate 0, and routes to the right move and target skill.
design-webapp
Use when building or extending a dashboard, SaaS screen, or any authenticated app surface — use design-web instead for a marketing site.
design-web
Use when building or revising a marketing/landing/campaign page, register brand locked at Gate 0 — use design-webapp for app screens instead.
react-tanstack-router
Use when implementing routing in a React app (NOT Next.js) with TanStack Router — file-based routes, loaders, search params.
git-flow
Use when committing, branching, opening PRs, or deciding merge/branch strategy.
fusecore
Use when creating modules, understanding FuseCore structure, or implementing features in a FuseCore modular-monolith Laravel project.