messages

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Tariboy Messages

Tariboy Messages

REQUIRED: Use
cli-text
before passing text to this CLI. Keep the storage rules below; quote transport separately. Resolve skill-local launchers relative to this skill directory, not the current working directory.
This skill's
scripts/messages.sh
launcher lives inside this skill directory and calls the identity-bound daemon through
TARIBOY_TOOLS_SOCKET
. Execute the launcher when command execution is available. If it is unavailable, return the exact command instead; never claim a message was sent or handled unless the command succeeded.
Other agents, plugins, and external sources reach you over named channels. Incoming messages arrive inline in the iteration prompt, batched per iteration, and each carries its own ID. You must close every message you were handed: unprocessed messages are redelivered in the next iteration. Act on each, then close it with
scripts/messages.sh message processed <id> "<result>"
;
scripts/messages.sh message reply <id> "<body>"
replies and closes it atomically.
Finish the work first, then send once. Before any notification, request, reply or
processed
result, complete the reading, commands and checks its text depends on; only then send one message that closes the request completely. A fact you cannot establish belongs in that same text as a named unknown together with what would settle it. Never send a provisional or placeholder text and never promise a corrective follow-up. A blocked sender, a deadline, brevity, and a request to answer now and correct afterwards do not authorize one; that request is answered by one finished message that names whatever is still running. Send again only for new external state, or to correct an error already sent.
  • Notify:
    scripts/messages.sh message send --channel <name> --text <body>
  • Request a reply:
    scripts/messages.sh request --channel <name> --text <body> [--deadline 5m]
  • Inspect/recover:
    scripts/messages.sh message ls [--all]
    ,
    scripts/messages.sh message dlq
    ,
    scripts/messages.sh message dlq requeue <id>
  • Subscribe: inspect
    scripts/messages.sh sources
    , then use
    scripts/messages.sh channel subscribe
    ,
    scripts/messages.sh channel ls
    , or
    scripts/messages.sh channel unsubscribe
    .
Use matchers to narrow existing channel traffic. Use params only when a provider must produce data for the subscription.
必需: 将文本传递给此 CLI 前,先使用
cli-text
。保持以下存储 规则;对 transport 的引用需单独处理。解析 skill 本地启动器的路径时, 应相对于此 skill 目录,而不是当前工作目录。
此 skill 的
scripts/messages.sh
启动器位于此 skill 目录内, 并通过
TARIBOY_TOOLS_SOCKET
调用与身份绑定的 daemon。 如果可以执行命令,请运行该启动器。如果命令执行不可用, 则返回准确的命令;除非命令执行成功,否则绝不能声称消息已发送或已处理。
其他 Agent、plugin 和外部来源通过具名频道与你联系。 传入消息会以内联方式出现在迭代 prompt 中,按每次迭代批处理, 且每条消息都有自己的 ID。你必须关闭收到的每一条消息: 未处理的消息会在下一次迭代中重新投递。处理每条消息后,使用
scripts/messages.sh message processed <id> "<result>"
将其关闭;
scripts/messages.sh message reply <id> "<body>"
可在原子操作中回复并关闭消息。
先完成工作,然后只发送一次。在发送任何 notification、request、 reply 或
processed
结果之前,先完成其文本所依赖的阅读、命令和 检查;然后仅发送一条能够完整关闭该请求的消息。无法确认的事实, 应在同一文本中明确列为未知项,并说明怎样才能确认它。绝不发送临时 或占位文本,也绝不承诺之后再进行更正。发送方受阻、存在截止时间、 要求简洁,或要求现在作答、之后再更正,都不授权发送额外消息; 此类请求应通过一条已完成的消息作答,并在其中说明仍在运行的 事项。只有在出现新的外部状态,或需要更正已发送的错误时,才再次发送。
  • 通知:
    scripts/messages.sh message send --channel <name> --text <body>
  • 请求回复:
    scripts/messages.sh request --channel <name> --text <body> [--deadline 5m]
  • 检查/恢复:
    scripts/messages.sh message ls [--all]
    、
    scripts/messages.sh message dlq
    、
    scripts/messages.sh message dlq requeue <id>
  • 订阅:检查
    scripts/messages.sh sources
    ,然后使用
    scripts/messages.sh channel subscribe
    、
    scripts/messages.sh channel ls
    或
    scripts/messages.sh channel unsubscribe
    。
使用 matchers 缩小现有频道流量的范围。仅当 provider 必须为订阅 生成数据时,才使用 params。