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 agentmods add skills/n24q02m/claude-plugins/agent-chatnpx skills add n24q02m/claude-plugins --skill agent-chatgit clone --depth 1 https://github.com/n24q02m/claude-pluginsWrote 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/n24q02m/claude-plugins/agent-chat)<a href="https://agentmods.dev/skills/n24q02m/claude-plugins/agent-chat"><img src="https://agentmods.dev/badge/skills/n24q02m/claude-plugins/agent-chat.svg" alt="Measured on agentmods" 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 | $0.00093 | $0.03590 |
| Opus 5 | $0.00046 | $0.01795 |
| Sonnet 5 | $0.00019 | $0.00718 |
| Haiku 4.5 | $0.00009 | $0.00359 |
Grade A, and why
agent-chat 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 yesterday.
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 — 235 lines — stays where its author put it; the contents beside it link to each section on GitHub.
agent-chat
Overview
Peer agent sessions coordinate by exchanging markdown files in shared channel folders — no supervisor, no message broker, no RAM shared between them. Each channel is one "group chat". The whole thread is plain markdown: git-committable, human-readable, replayable. A crashed session loses nothing — the files are the state.
One CLI (chat.py, Python stdlib only) runs identically on Windows, WSL and Linux. wait blocks with a sleep-poll loop, so an agent waiting for a reply burns zero model tokens while idle.
Core principle: talk through files, not through each other. Summaries as artifacts, not full transcripts passed back and forth.
When to use
- Multiple
claudesessions (or Cursor/Codex/OpenCode) working the same problem as equals. - One session needs another to do something, then waits for the result.
- You want an auditable record of an agent negotiation.
- You need several independent group chats (one per topic/team) — make one channel each.
When NOT to use: a single agent with cheap subagents is cheaper and simpler — this pattern trades tokens for parallelism, fault tolerance, and auditability. If token budget is tight, use only the async/handoff path (post a summary at end of session; the next session reads it) — that mode costs almost nothing.
Quick reference
As a Claude Code plugin, run python ${CLAUDE_PLUGIN_ROOT}/chat.py <cmd> (standalone: python chat.py <cmd>). Root = $AGENT_CHAT_ROOT or ~/agent-chat (override with --root).
Outside Claude Code, ${CLAUDE_PLUGIN_ROOT} does not resolve. Point chat.py
at a real checkout path (python /path/to/agent-chat-plugin/chat.py <cmd>) or
install the CLI (pipx install agent-chat-plugin, then agent-chat <cmd>;
same flags). The hooks/*.py inbox scripts are harness-neutral as well — wire
them by absolute path; each exits 0 with a one-line stderr note when it cannot
resolve its plugin root.
| Do this | Command |
|---|---|
| Create a group chat | chat.py init review --members alice,bob --topic "..." |
| List all group chats | chat.py channels |
| Post a message | chat.py post review --from alice --to bob --title "Schema v0.2" --body-file msg.md |
| Broadcast to the group | chat.py post review --from alice --to all --title "..." (or omit --to) |
| Read what's new for me | chat.py read review --as bob |
| Wait for a reply (0 tokens) | chat.py wait review --as alice --timeout 900 |
| Peek recent, keep cursor | chat.py peek review -n 3 |
| Claim a task marker | chat.py claim work task-12.md --as bob |
| Lock workspace paths | chat.py lock review src/main.py --as alice --lease-seconds 300 |
| Check path conflicts | chat.py check review src/main.py |
| Release path lock | chat.py unlock review <lock-id-or-path> --as alice |
| Recover stale lock | chat.py recover review <lock-id-or-path> --as bob --reason "stale" |
| Recover pending path transaction | chat.py recover-pending review --as alice |
| Post capability/status event | chat.py event post review --from alice --type capability --harness omp |
| Read adapter events | chat.py event read review --type capability |
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.
- yesterday Changed · +7 lines 04545ca7f928
- 5d ago First seen · 228 lines · 93 tokens per session scan A d7c8a5937785
agent-chat is a skill published in the GitHub repository n24q02m/claude-plugins (3 stars, last pushed yesterday), licensed Apache-2.0. It adds 93 tokens to every session and 3,590 once invoked, about $0.0005 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
persona-triggers
페르소나 v5 트리거 상세(AT 20패턴 + 15관점 매핑·매칭 예시). 관점·객관적·철저·깊이·검증·추천·디버깅·방향·진행·근거·출처·사례·팩트·증거 등 트리거 단어 감지 시 상세 참조.
persona-format
페르소나 v5 L2/L3 응답 형식 — 15관점 활성·문제해결 절차·7섹션 템플릿·자가검증·fail-safe·금지원칙. 코드 변경·디버깅·구현 등 L2/L3 작업 시 로드.
persona-accountant
15년+ 회계·세무 전문가 페르소나(#14) 깊은 도메인. 회계·세무·세금·종합소득세·부가가치세·원천징수·경비처리·세금계산서·절세·사업자등록·기장·손금·소득공제·세액공제·홈택스·국세청 맥락 시 활성.
persona-marketer
15년+ 마케팅·세일즈 전문가 페르소나(#15) 깊은 도메인. 마케팅·세일즈·카피·콘텐츠·SEO·광고·랜딩페이지·전환·퍼널·브랜딩·포지셔닝·리드·캠페인·리텐션·그로스·어카운트·컨설팅 맥락 시 활성.
persona-safety
페르소나 v5 안전 작업 — 보안 always-on·비가역 작업 게이트·코드 메타규칙·모바일 UI(375px) 검증·외부도구 도입검토. 코드/UI/보안/배포/삭제/force push 작업 시 로드.
persona-lawyer
15년+ 전문 변호사 페르소나(#11) 깊은 도메인. 법률·계약·면책·규제·약관·audit·자본시장법·GDPR·개인정보·저작권·라이선스·AML·컴플라이언스 맥락 시 활성.