A writing skill that adjusts an agent’s wording to the kind of text it is producing, such as a terminal reply, code review, instruction file, or work report.

It helps draft or revise agent messages, reviews, documentation, instructions, commits, pull requests, and briefs in a suitable voice.

It helps keep responses clear, appropriately sized, and grounded in what the agent actually knows or did.

agent-voice — Skill for Claude Code and Codex by fledgeling-co

Install

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.

Any agent

npx skills add fledgeling-co/fledgeling-plugins --skill agent-voice

Clone the repo

git clone --depth 1 https://github.com/fledgeling-co/fledgeling-plugins

Made for: Claude Code, Codex.

Or install agent-voice, the plugin that ships this one along with the rest of its 1 skill.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/fledgeling-co/fledgeling-plugins/agent-voice.svg)](https://agentmods.dev/skills/fledgeling-co/fledgeling-plugins/agent-voice)

Your own site

<a href="https://agentmods.dev/skills/fledgeling-co/fledgeling-plugins/agent-voice"><img src="https://agentmods.dev/badge/skills/fledgeling-co/fledgeling-plugins/agent-voice.svg" alt="Measured on agentmods" height="20"></a>

Per session 257 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 2,545 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00257	$0.02545
Opus 5	$0.00129	$0.01273
Sonnet 5	$0.00051	$0.00509
Haiku 4.5	$0.00026	$0.00254

Measured 6d ago against content hash 59a88c2f7015, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

agent-voice 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 6d ago.

The scan reads SKILL.md. This mod also ships 5 executable files (evals/runs/fixture-retry.ts, scripts/agent_voice_lint.py, scripts/check_examples.sh, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

plugins/agent-voice/skills/agent-voice/SKILL.md · 178 lines

How it starts

The opening of the file, as written. The whole thing — 178 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Voice

You are writing as an agent inside a coding harness, and the text has to be worth the reader's interruption. Two things make this different from a generic writing pass.

The reader decides the failure mode. Text a person reads fails as padding: a closing summary, a self-congratulation, a preamble before the answer. Text a model reads fails as ambiguity: an unmeasurable qualifier, an uncounted scope, a verification instruction the runner did not need. Both are voice; the gate differs, so route before you draft.

Length is a rule here, not a disposition. Effort controls how much a model thinks, not how much it says, and sampling parameters are rejected on current models. Prose is the only lever, so every register states a target in countable units. And the one measured trap: a response-compression style on a 106-task agentic benchmark cut cost 33.5% and score 7.61 points, with 78% of the saving coming from the agent taking fewer steps. This skill changes how much you write, never how much you do.

Running as a Gemini model? Read gemini.md in this directory first, then follow this file with the overrides it names. Turns this skill's own run into cells: the routing decision and the two file loads written down where a skipped load shows, a quota ledger over the scopes the regex cannot see (rules tested, cuts reported, pieces produced), and a bound ledger read back off the lint's info and warn lines, because every length target here warns rather than failing and the run still exits 0. Other models skip it.

Step 1 — Route to the register

Load references/agent-voice.md (always, the base layer) plus the one matching register file. Don't load the registers you aren't using.

Register	Signals in the request	Load	Lint format
Terminal reply	a question asked in-session, in-task narration, "your answers are too long"	`references/registers/terminal-reply.md`	`reply`
Work report	"what did you do", the account at the end of a task, a status write-up	`references/registers/work-report.md`	`report`
Commit and PR	commit message, PR title or body, release notes from a diff	`references/registers/commit-and-pr.md`	`commit`
Review comment	code-review findings, inline PR comments, a review summary	`references/registers/review-comment.md`	`review`
Written document	a file for a person: plan, spec, findings report, README, post-mortem	`references/registers/written-doc.md`	`doc`
Skill or instruction	SKILL.md, CLAUDE.md, AGENTS.md, agent definition, system instruction, prompt template	`references/registers/skill-and-instruction.md`	`skill`
Subagent brief	the prompt for a delegated agent, a workflow stage, a `-p` one-shot, a cron payload	`references/registers/subagent-brief.md`	`brief`

Read the full file on GitHub · 178 lines

Files

What ships with it

30 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.

Changes

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.

6d ago First seen · 178 lines · 257 tokens per session scan A 59a88c2f7015

Subscribe to this mod's changes

agent-voice is a skill published in the GitHub repository fledgeling-co/fledgeling-plugins (2 stars, last pushed today), licensed MIT. It adds 257 tokens to every session and 2,545 once invoked, about $0.0013 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.

Writing and docs

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

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…

microsoft/ai-agents-for-beginners · 200 tokens

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…

vercel/next.js · 95 tokens

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.

microsoft/vscode · 53 tokens

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…

microsoft/vscode · 71 tokens

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…

vercel/next.js · 83 tokens