Track and surface sub-agent dispatch calibration ratios. Extract from JSONL transcript or read existing ledger. Per-agent clean/blocked/errored/partial trends.
中文简介正在准备中。当前显示原文说明。
Mute a Caddy hint class for a window. Defaults to 4h auto-expiry. Classes — memory, context, body, routing, all.
中文简介正在准备中。当前显示原文说明。
实测评分 未实测
查看安装方法安装前可以查看交给 AI 代理的指令内容。
Suppress a Caddy hint class for a bounded window. State lives in memory/CADDY_MUTES.json. daemons/caddy.sh reads it on every UserPromptSubmit and skips matching class hints when active.
memory — [CADDY:memory] MEMDB, DIGEST hintscontext — [CADDY:context] CTX, CAPSULE hintsbody — [CADDY:body] BODY-CHECK hintrouting — [CADDY:routing] ROUTE, /orient, STYLE, DICT hintsall — nuclear; mutes every class/caddy-mute --class memory # 4h default
/caddy-mute --class memory --hours 1 # 1h
/caddy-mute --class memory --forever # explicit permanent (rare)
/caddy-mute --class all # nuclear
/caddy-mute --class memory --reason "design session, hint fatigue"
If no --hours and no --forever: default 4 hours.
Compute muted_until and Edit memory/CADDY_MUTES.json:
{
"memory": {
"muted_until": "2026-05-03T18:00:00Z",
"reason": "design session, hint fatigue",
"default_duration_hours": 4
}
}
For --forever: set muted_until to 9999-12-31T23:59:59Z. Surface to the operator when setting forever ("muted memory class indefinitely — confirm?") so it doesn't drift into permanent silence by accident.
For --class all: write a single all block; caddy.sh class_muted checks all in addition to the requested class.
To unmute early, Edit memory/CADDY_MUTES.json and remove the class entry, or set muted_until to the past. There's no /caddy-unmute skill yet — the entry self-expires at muted_until, which is the point.
all for >24h. If you need all muted longer than that, the hint surface itself is broken and Caddy needs editing, not silencing.Spec: [[concepts/Somatic v0.4.1 Wiring]] §6. Class taxonomy and 4h default were Codex+aigent-OS joint design decisions in the 2026-05-02 session.
还没有评价。欢迎分享你的使用感受。
无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。
还没有评论。来写第一条吧。
Track and surface sub-agent dispatch calibration ratios. Extract from JSONL transcript or read existing ledger. Per-agent clean/blocked/errored/partial trends.
中文简介正在准备中。当前显示原文说明。
Compose the AIgent's body state by reading existing signals. Lazy compute. Never polled. Reports the five pressures + recommended reflex.
中文简介正在准备中。当前显示原文说明。
Generate a structured delegation brief for any task
中文简介正在准备中。当前显示原文说明。
Quarterly skill that walks the skill-index against the actual SKILL.md files on disk, reports drift (skills indexed that no longer exist, skills on disk that aren't enrolled, skills whose descriptions have changed since enrollment). Hardens the auto-enrollment pipeline against silent rot.
中文简介正在准备中。当前显示原文说明。
When a [CADDY] hint fires and the principal is confused why, /caddy-explain walks the scoring transparently — which trigger matched, what points it earned, what the threshold was, what other skills were close. Pairs with the legibility thesis. Caddy's decisions should be inspectable on demand.
中文简介正在准备中。当前显示原文说明。
Enroll a skill into Caddy's index so it can be surfaced when the right prompt arrives. Use when a new skill is created, copied in, or when a skill needs updated triggers. Usage — "/caddy-enroll <path to SKILL.md>".
中文简介正在准备中。当前显示原文说明。