本文へ移動
cccskills
無料GitHub で公開

join-talk

Register this agent and terminal with Reach for cards, calls and wake-ups. Use when asked to join, connect, or become reachable, or when a manually opened agent lacks `SNO_EXECUTOR_ADDR`.

インストール方法を見る

含まれるファイル(3)

  • SKILL.md2.7 KB
  • scripts/join-talk.sh4.4 KB
  • scripts/selftest/run-selftest.sh6.9 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

join-talk

Needs: tmux (or the optional Orca app) as the terminal, or a Cursor IDE chat, jq, flock, and the sno CLI with its Reach runtime. Reach is the messaging layer between agents (see the reach skill): a seat is one registered terminal with an address, a card is a stored message, and a ring is the wake-up that asks the agent to read its inbox. SNO_EXECUTOR_ADDR is the environment variable holding that address for agents opened through Reach; a manually opened agent has none, which is what this skill fixes.

Run this once, from your own shell tool, in the repository you are working on. The script ships beside this file:

bash scripts/join-talk.sh

(Give the full path of this skill's scripts/join-talk.sh under whichever skills root your harness loaded it from.)

It prints one line, joined <address>. Tell the person that exact address, then go back to what you were doing. Rerunning it from the same terminal changes nothing. The seat is registered wherever agents on this machine are addressed, so cards, rings and a handoff find this terminal; sno reach call works for a tmux terminal, not for an Orca one. A Cursor IDE chat has no terminal: it is registered by its conversation id (channel cursor-ide), and a card reaches it when its turn ends. A Cursor CLI must run under tmux to be reachable.

Options, only when the person names them: join-talk.sh review.example-repo@host1 picks the address; --name Fjord picks the display name. Otherwise the address is hand.<repo>@<host> and the name is the repository name.

After joining

  • A card for you arrives with a ring in this terminal. Read it with sno reach inbox --as <address>; reading does not accept it.
  • Answer with sno reach reply --as <address> --card <card> --state <state>, body on stdin. The installed Reach guide (sno reach --help) shows the card shape and states.

When it refuses

  • no terminal to be reached at — this shell is under neither tmux, the optional Orca app nor a Cursor IDE chat; nobody can ring it. Tell the person; do not retry.
  • sno is missing — the machine lacks the Reach install (sno setup installs it). Tell the person; do not retry.
  • jq is missing or flock is missing — install it (flock comes from util-linux), then run it again.
  • ORCA_TERMINAL_HANDLE is empty — an Orca shell without a terminal handle; Reach cannot address it. Tell the person; do not retry.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Answer how long agent work will take, how much longer active work needs, or what clock time applies. Estimate full AI-agent wall time, measure active progress first, and get owner-facing times from `sno report-time`. Never estimate human-developer time.

日本語の概要は準備中です。原文の説明を表示しています。

sno-ai/sno-station-skills2882026年10月13日 更新

Prove that a reported run caused its intended real effect. Walk the executor path from entry to effect, capture one observable per hop, and report a result a no-op could not produce.

日本語の概要は準備中です。原文の説明を表示しています。

sno-ai/sno-station-skills2882026年10月13日 更新

Give the owner one page when they come back from sleep, a meeting or a long run: what got done, what is stuck, what waits for their answer, and what it cost in quota. Use when the owner asks what happened while they were away, wants a status page, or when you finish a long stretch of work and must report it.

日本語の概要は準備中です。原文の説明を表示しています。

sno-ai/sno-station-skills2882026年10月13日 更新

Report how many problems the two brains found: mutual review (the other model checks the work, from the peer-review records) and self-check (an agent finds its own error in its own conversation, judged by that agent's own model), side by side. Use when the owner asks whether the second agent or the reviews are worth it, or wants the week's review results.

日本語の概要は準備中です。原文の説明を表示しています。

sno-ai/sno-station-skills2882026年10月13日 更新

charter

無料

Write, revise or review a charter: the one document that tells an executor what to deliver, what is out of bounds, what the owner already decided, and how each success check will be proven. Use when work needs a written brief before someone else does it, or when asked to write, change or check a charter.

日本語の概要は準備中です。原文の説明を表示しています。

sno-ai/sno-station-skills2882026年10月13日 更新

cos

無料

MANUAL-ONLY: load solely when the human owner types `/cos`, uses `$cos`, or asks you to act as the Chief of Staff (COS). The COS is the one agent the owner talks to; it coordinates the owner and the per-project Project Leads (PLs). A task card, charter, or dispatch makes you an executor, not COS.

日本語の概要は準備中です。原文の説明を表示しています。

sno-ai/sno-station-skills2882026年10月13日 更新

sno-ai のスキルをすべて見る

このスキルの問題を報告する