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 synthesisengineering/synthesis-skills --skill synthesis-agent-correspondencegit clone --depth 1 https://github.com/synthesisengineering/synthesis-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/synthesisengineering/synthesis-skills/synthesis-agent-correspondence)<a href="https://agentmods.dev/skills/synthesisengineering/synthesis-skills/synthesis-agent-correspondence"><img src="https://agentmods.dev/badge/skills/synthesisengineering/synthesis-skills/synthesis-agent-correspondence/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/synthesisengineering/synthesis-skills/synthesis-agent-correspondence"><img src="https://agentmods.dev/badge/skills/synthesisengineering/synthesis-skills/synthesis-agent-correspondence.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00115 | $0.05705 |
| Opus 5 | $0.00057 | $0.02853 |
| Sonnet 5 | $0.00023 | $0.01141 |
| Haiku 4.5 | $0.00012 | $0.00571 |
Grade A, and why
synthesis-agent-correspondence 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 12d 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 — 290 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Correspondence
The core principle
Ghostwriters and executive assistants have drafted correspondence for the people they serve for as long as there has been correspondence to draft. Senior executives have long run their correspondence through personal staff at several levels of involvement at once: some messages the principal writes personally; some the principal dictates and staff transmit; some the principal directs at a high level ("reply warmly, decline the date") and staff compose in the principal's voice; and some the staff handle entirely, with the principal's knowledge of the system rather than of the message. None of that was ever considered dishonest, and none of it carried a disclaimer — because the principal owned the relationship and the direction, and the staff were a trained extension of the principal's judgment.
AI agents change two things about that old system, and only two. First, some platforms stamp agent-performed sends visibly, so pretending is not even an option. Second, an AI agent is not yet as reliable as a trained human staff — a hallucinated detail attributed to the principal's own hand damages trust in everything the principal actually wrote. So unlike the human-staff era, disclosure is warranted. The design question is what the disclosure should track.
Disclosure should answer the one question the recipient actually cares about: whose words are these? Everything else — how the sausage was made, what approval workflow ran — is internal governance, not disclosure.
The two questions, and the three lanes
Every outgoing message answers two questions: whose words are these, and who performed the send. Those two answers sort all correspondence into three lanes on a single axis — how much of the principal is in the words:
| Lane | Whose words | Who sends | Recipient-facing disclosure |
|---|---|---|---|
| Principal-direct | The principal's | The principal | None. Nothing to disclose, on any channel — regardless of how much AI research, drafting, or polish went into it. The principal read every word and performed the send; that is the ghostwritten letter the principal signs. |
| Assistant lane | The principal's — composed, dictated, or edited to the point of genuine ownership | The agent | A single authorship signature: the principal wrote this, working through the named agent. One line, one meaning, no variants. |
| Bot lane | The agent's, under the principal's direction — per-message instruction or standing rules | The agent | The persona's own signature: the named agent introduces itself, states the principal's direction, and promises the principal reads every reply. Wording may reflect how deep the direction ran. The body speaks in the persona's own voice (see the voice axis below). |
What ships with it
3 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.
- 12d ago First seen · 290 lines · 115 tokens per session scan A 741646793745
synthesis-agent-correspondence is a skill published in the GitHub repository synthesisengineering/synthesis-skills (19 stars, last pushed today), licensed Apache-2.0. It adds 115 tokens to every session and 5,705 once invoked, about $0.0006 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 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…