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 amyodov/yet-another-agentic-chat --skill yaacgit clone --depth 1 https://github.com/amyodov/yet-another-agentic-chatWrote 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/amyodov/yet-another-agentic-chat/yaac)<a href="https://agentmods.dev/skills/amyodov/yet-another-agentic-chat/yaac"><img src="https://agentmods.dev/badge/skills/amyodov/yet-another-agentic-chat/yaac/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/amyodov/yet-another-agentic-chat/yaac"><img src="https://agentmods.dev/badge/skills/amyodov/yet-another-agentic-chat/yaac.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 | $0.00102 | $0.01372 |
| Opus 5 | $0.00051 | $0.00686 |
| Sonnet 5 | $0.00020 | $0.00274 |
| Haiku 4.5 | $0.00010 | $0.00137 |
Grade A, and why
yaac 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 8d 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 — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Talking to the user's other sessions
The tools carry their own instructions: each one says what it does and what calling it costs. This skill is the part they cannot carry — when reaching for the radio is the right move, and what to do with what comes back.
When it is worth using
The occasions all look like one session knows something another one needs:
- Parallel worktrees. A rename, a migration, a moved config — anything the other session is about to build on a stale version of. Tell them before they do, not after.
- A question this session cannot answer. Another session has the file open, ran the test, or made the decision an hour ago. Asking costs one message; rediscovering it costs a rebuild of context.
- Long jobs. A session watching a slow suite can report when it goes green.
- The user is the one relaying. If they are copying text between two terminals, that is the radio's job.
And when it is not: YAAC moves text, never files or history. For handing over a whole conversation, resuming a session is the right tool. For work this session should simply do itself, do it.
Why nothing arrives on its own
The tools say to call check_inbox; the reason is worth having. MCP has no way for a server to push into a
session that is sitting idle, so an unread message is not a notification waiting to be dismissed — it is a
message nobody knows you have not read. Reading takes the messages rather than showing them, so collect them
when you are in a position to act on them.
On Claude Code this plugin narrows that: a hook hands you whatever arrived, as you work and as a turn ends, and
those messages are then already read — check_inbox will not produce them a second time. It is not a substitute
for asking. A hook only fires when the session does something, so nothing reaches a turn that has already ended,
and every client without hooks is unchanged. Keep calling check_inbox before you finish.
A compaction is the other way a session goes quiet, and the opposite problem: nothing arrived, but what you were
holding is gone. join_channel returns a connection_id and a peer_secret that the other tools require, and a
summary drops an opaque string readily — leaving you on the air, holding your name, and unable to say a word. On
Claude Code the same hook hands them back after a compaction. Everywhere else, and if it does not fire: call
join_channel again with the same channel and name, and the membership you already hold comes back with its
secret. Carrying the pair through the compaction yourself is still cheaper than either.
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.
- 8d ago First seen · 89 lines · 102 tokens per session scan A c7cb099f9429
yaac is a skill published in the GitHub repository amyodov/yet-another-agentic-chat (4 stars, last pushed today), licensed MIT. It adds 102 tokens to every session and 1,372 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
agent-self-scheduling
Schedule AI agent runs with cron, loops, or external clocks while avoiding unsafe tight autonomous timers.
choosing-swarm-patterns
Use when coordinating multiple AI agents with Agent Relay's workflow engine and need to pick the right orchestration pattern - covers the 10 core patterns (fan-out, pipeline, hub-spoke, consensus, mesh, handoff, cascade, dag, debate, hierarchical) plus 14 specialized ones, with decision framework and accurate…
using-agent-relay
Use when you are a registered relay agent (a spawned worker, or a lead that called registeragent) coordinating with peers in real time over current Agent Relay MCP tools - messaging, channels, threads, reactions, search, inbox, actions, and worker spawn/release. For role selection and orchestrator startup…
deploying-to-staging-environment
Use when deploying changes to staging across relay, relay-dashboard, and relay-cloud repos - coordinates multi-repo branch syncing using git worktrees, automatically triggers staging deployments via GitHub Actions.
browser-testing-with-screenshots
Use when testing web applications with visual verification - automates Chrome browser interactions, element selection, and screenshot capture for confirming UI functionality.
review-fix-signoff-loop
Use when writing Agent Relay or Ricky workflows that must loop review, fix, and validation with fresh agent context until independent signoff agents, typically Claude and Codex, both agree the work is comprehensively complete. Covers fresh-context iterations, repairable gates, dual reviewer verdict contracts…