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。
List every installed skill, when it was last invoked, how often, and whether the description still matches the trigger patterns the principal actually uses. Flags dead skills for retirement and frequently-needed-but-missing capabilities for skill creation. Pairs with Caddy.
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。
Auditing the skill catalog itself. As the install grows from 5 skills to 50, the catalog goes from "I remember every skill" to "I have no idea what's in here." This skill makes the catalog legible to its owner.
A markdown report with four sections:
| Skill | Last used | Total invocations (90d) | Description |
|---|---|---|---|
| /open | today | 47 | Start session, load vault context |
| /close | today | 47 | End session, commit memory |
| /caddy-enroll | 5 days ago | 12 | Add a skill to Caddy index |
| /skill-name | 23 days ago | 3 | Example skill entry |
| /old-experiment | 89 days ago | 1 | DEAD — flagged for retirement |
These are candidates for retirement. List with:
For each skill, compare the trigger keywords in .claude/skill-index.json against the actual prompt patterns that preceded each invocation. Flag mismatches:
⚠️ /channel-audit
Triggers: "channel status", "YPP progress", "kill a channel"
Actual prompts that led to invocation: "how are channels doing", "performance check", "review channels"
Suggested triggers to ADD: "performance check", "review channels", "how are channels doing"
Heuristic: scan recent prompts (last 30 days) where Caddy fired no hint and the principal manually invoked Read/Bash/Grep instead of a skill. Cluster by keywords. If a cluster has 3+ instances, flag as a capability gap.
🆕 Possible missing skill: "weekly metrics rollup"
Inferred from prompts: "weekly metrics", "this week's numbers", "rollup the week"
Manual tools used: 4 Bash calls per occurrence
Suggestion: codify as /weekly-rollup skill via /skills-builder
vault/daily/, ~/.claude/projects/, hook output logs)vault/memory/SKILL_AUDIT_<YYYY-MM-DD>.md for trend trackingThe Caddy pattern only works if the catalog stays useful. Skills accumulate faster than they're retired. Trigger patterns drift from the principal's actual phrasing. Capability gaps go unnoticed for months. Skill audit is the corrective loop.
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。