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 yohey-w/kagemusha --skill meeting-boardgit clone --depth 1 https://github.com/yohey-w/kagemushaWrote 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/yohey-w/kagemusha/meeting-board)<a href="https://agentmods.dev/skills/yohey-w/kagemusha/meeting-board"><img src="https://agentmods.dev/badge/skills/yohey-w/kagemusha/meeting-board/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/yohey-w/kagemusha/meeting-board"><img src="https://agentmods.dev/badge/skills/yohey-w/kagemusha/meeting-board.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 | $0.00453 | $0.02999 |
| Opus 5.5 | $0.00181 | $0.01200 |
| Sonnet 5.5 | $0.00091 | $0.00600 |
| Haiku 4.5 | $0.00045 | $0.00300 |
Grade A, and why
meeting-board 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 today.
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 — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
meeting-board
Meet の字幕を、話者つきで整理して別窓に出す(手順と使い方は README.md)
Meet のタブ ── 拡張機能(content.js)── background.js ──▶ ヘルパー(helper/server.js・127.0.0.1:47431)
字幕の文字だけ読む │ 字幕がたまったら claude -p へ
▼
別窓のボード(http://127.0.0.1:47431/)
エージェントとして呼ばれたら — 完了の形と止まる場面
頼まれやすいのは、会議用 PC への導入の手伝い・会議の前の動作確認・「字幕が出ない」の切り分け・Meet の画面変更への追従。
完了の形:
node templates/skills/meeting-board/tests/run.jsが全部通る(Node 18 以上のどれでも同じ。外部ライブラリなし・本物のclaudeは呼ばない)- 動作確認は
node helper/server.js --replay sample/sample_transcript.txt --notes sample/sample_notes.md --onceが、next_action(次の一手)・checklist(stateがdone/deferred/open)・topicsのある JSON だけを標準出力に出して終了コード 0(これは手元のclaudeを 1 回呼ぶ。架空の会議なので送ってよい) 画面は、偽 claude(CLAUDE_BIN=tests/fixtures/claude_stub.js・STUB_BOARD=tests/fixtures/board_v2_sample.json)で--replayを立て、http://127.0.0.1:<port>/(会議中)と/?view=after(会議の前と後)を見る - 拡張機能は人が
chrome://extensionsで読み込む。エージェントは「拡張機能が動いた」と書かない——Meet のページを実機で見たのは人だけ
止まって人に聞く場面:
- 🔴 待ち受けを
127.0.0.1以外にする前・外へ出る通信(claude以外)を足す前・manifest.jsonの権限やhost_permissionsを増やす前 - 🔴
data/の中身(実際の会議の逐語)を読む・貼る前。診断(ボードの「診断をコピー」・GET /diag)は発言を含まない作りなので、こちらを頼む - 🔴
helper/prompt.md・helper/prompt_patch.mdを変える前(出力の形がボードと試験の前提になっている。変えるならlib/board.js・board-core.js・board.html・試験も揃える。lib/extract.jsのnormalizeBoard()は 0.1 の形の土台なので変えない) - 🔴 議事録の素(
--minutes・「会議を終える」のminutes.md)を、ファイルと標準出力以外へ送る仕組みを足す前(メール・チャットなどへの自動送信はしない作り) - 🔴
claudeの安全のための設定(helper/lib/claude.jsのFLAGS_FULL)を外す・外して呼び直す仕組みを足す前(古い版では止めてclaude updateを案内する作り)
Meet の画面が変わって字幕が読めないとき
読み取りの要所は extension/content.js の locate()(字幕の領域を探す 5 つの手順)と findBlocks()(話者ブロックを探す)。診断の tree は、字幕の領域の下を 3 段、タグ・クラス名・子の数・文字数(長さだけ)で写してある。診断 1 つあれば、実際の字幕の構造を再現した固定ページを作って、content.js を直せる(本文は要らない)。現行の Meet は、字幕の領域が [role="region"] で aria-label が「字幕」、子が 2 つ(字幕の入れ物と「最新の字幕に移動」ボタン)。
What ships with it
60 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.
- .gitattributes 209 B
- data/.gitignore 14 B
- extension/background.js 5.8 KB runs code
- extension/captions-core.js 10 KB runs code
- extension/content.js 19 KB runs code
- extension/manifest.json 613 B
- helper/board-core.js 17 KB runs code
- helper/board.html 49 KB
- helper/lib/app.js 38 KB runs code
- helper/lib/args.js 6.1 KB runs code
- helper/lib/board.js 21 KB runs code
- helper/lib/claude.js 12 KB runs code
- helper/lib/diag.js 3.6 KB runs code
- helper/lib/extract.js 5.5 KB runs code
- helper/lib/httpserver.js 8.7 KB runs code
- helper/lib/importers.js 6.4 KB runs code
- helper/lib/minutes.js 5.8 KB runs code
- helper/lib/notes.js 5.5 KB runs code
- helper/lib/prompt.js 6.5 KB runs code
- helper/lib/replay.js 6.4 KB runs code
- helper/lib/script.js 9.3 KB runs code
- helper/lib/store.js 3.8 KB runs code
- helper/lib/transcript.js 6.5 KB runs code
- helper/lib/util.js 1.4 KB runs code
- helper/messages/failed.txt 283 B
- helper/messages/no-node.txt 519 B
- helper/messages/old-node.txt 353 B
- helper/prompt_live.md 1.5 KB
- helper/prompt_patch.md 2.2 KB
- helper/prompt.md 9.5 KB
- helper/server.js 14 KB runs code
- README.md 18 KB
- sample/sample_notes.md 955 B
- sample/sample_transcript.txt 5.8 KB
- start.cmd 1.2 KB runs code
- start.sh 614 B runs code
- tests/board_core.test.js 15 KB runs code
- tests/board_v2.test.js 20 KB runs code
- tests/captions_core.test.js 10 KB runs code
- tests/claude.test.js 11 KB runs code
- tests/e2e/meet_fixture.html 6.4 KB
- tests/e2e/run_e2e.js 15 KB runs code
- tests/eval_score.test.js 2.2 KB runs code
- tests/eval/score_board.js 5.0 KB runs code
- tests/extract.test.js 4.5 KB runs code
- tests/fixtures/board_v2_sample.json 3.6 KB
- tests/fixtures/claude_stub.js 4.2 KB runs code
- tests/fixtures/sample_script.md 1.1 KB
- tests/fixtures/sample_teams.txt 502 B
- tests/fixtures/sample_teams.vtt 891 B
- tests/fixtures/sample_zoom.vtt 325 B
- tests/fixtures/support.js 3.8 KB runs code
- tests/importers.test.js 4.4 KB runs code
- tests/minutes.test.js 6.4 KB runs code
- tests/notes.test.js 3.3 KB runs code
- tests/prompt.test.js 19 KB runs code
- tests/replay_cli.test.js 11 KB runs code
- tests/run.js 1.5 KB runs code
- tests/script.test.js 11 KB runs code
- tests/server.test.js 56 KB runs code
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.
- today Changed · +15 lines · +127 tokens per session 36544725830c
- 3d ago First seen · 52 lines · 326 tokens per session scan A 4ec412c2ea4b
meeting-board is a skill published in the GitHub repository yohey-w/kagemusha (16 stars, last pushed today), licensed MIT. It adds 453 tokens to every session and 2,999 once invoked, about $0.0018 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 2026-10-21.
Other skills, from other repositories
batch
Batch job-application filler. Given several job URLs (or a urls.txt file), run the assisted-apply flow for each in its own browser tab — fit-gate, fill-or-flag, stop before Submit — skipping stale or low-fit postings, then present one combined review table so the human can approve each tab one by one. Use for /batch…
website-change-monitor
Watch important pages and notify when content changes.
slack
Interact with Slack workspaces using browser automation. Use when the user needs to check unread channels, navigate Slack, send messages, extract data, find information, search conversations, or automate any Slack task. Triggers include "check my Slack", "what channels have unreads", "send a message to", "search Slack…
browser-use
Direct browser control via CDP for web interaction: automation, scraping, testing, screenshots, and site/app work. Also covers hosted cloud browsers for AI agents: the browser-use Python library (Browser(usecloud=True)), Playwright, Puppeteer or any CDP client.
cli-anything-macrocli
Use when the agent wants to define, list, inspect, or execute GUI macros via the MacroCLI CLI. Macros are parameterized, CLI-callable workflows — the agent invokes macro run and the system handles backend routing (plugin, file transform, accessibility, compiled GUI replay).
agent-browser
Use the agent-browser CLI to exercise web interfaces, inspect rendered accessibility state, verify interactions, and capture screenshots. Use for browser-based UI testing and evidence collection. Do not use as a substitute for deterministic Playwright coverage when an automated regression test is the requested…