能力

Kemetra2026-10-26查看 1864复制 1351

展示此工具包（kit）的只读能力清单——按生命周期状态、权限、要求和来源列出每项能力。当有人询问该工具包能做什么，或在没有数据库的情况下哪些功能可用时使用。不写入任何内容，也不输出评分。

📘如何使用这个技能

1
复制技能内容·点击上方「Copy SKILL.md」复制全部内容，或 Download 下载到本地
2
打开你的 AI 客户端·支持 Claude Code、Trae、Cursor、ChatGPT、DeepSeek、Kimi 等任意 AI 助手
3
粘贴为系统提示/知识文件·Claude Code：放入 .claude/skills/ 目录；普通对话：作为首条消息发送并说「请严格按这份技能文档执行」
4
直接提出你的需求·AI 会按技能定义的工作流、格式和约束来处理，无需再解释背景

💡 提示：技能生效后，每次对话只需一句需求，AI 会按 SKILL.md 的流程执行。

SKILL.md2313 chars

---
name: capabilities
description: >-
  Show the read-only capability inventory for this kit -- each capability by lifecycle state, authority, requirements and provenance. Use when someone asks what the kit can do or what works without a database. Writes nothing and emits no score.
---

# capabilities

The read-only capability inventory: one truthful, categorical read of
everything Seshat BI can do, derived from committed, reviewable metadata
(never from a file's mere existence).

## What it does

Reads the committed capability manifest (`docs/capabilities/capabilities.yaml`)
and reconciles it against the feeders that already own each fact -- the
`_DISPATCH` command table, `.claude/skills/*/SKILL.md` frontmatter,
`.seshat/kit-source.yaml` verbs, `docs/roadmap/roadmap.md` F-numbered ship
status, and `docs/quality/status-claims.yaml` -- then renders:

- a grouped human-readable read (the default), or
- a stable, deterministic JSON machine form (`--format json`).

## How to invoke it

Run the module directly from the repo root:

```
python -m seshat.capability_inventory
python -m seshat.capability_inventory --format json
```

There is no `retail capabilities` / `seshat capabilities` CLI verb -- this is a
deliberate, ratified choice (`docs/roadmap/decisions/cli-verbs-vs-skill-driven.md`,
Option B). Do not add one; do not touch `src/seshat/cli/parser.py` or
`_DISPATCH` to "wire this up". Simply run the module command above and relay
its output.

## What it is NOT

- **Not a gate.** It adds no `seshat check` rule; its presence/absence never
  blocks anything.
- **Not a readiness surface.** It reads no `readiness-status.yaml`, moves no
  stage, grants no approval. For per-table readiness use `seshat status`; for
  the next allowed action use `seshat next`; for repo drift use
  `seshat doctor`; for the governance gate use `seshat check`.
- **Not a scoring surface.** No maturity/confidence/completeness/health value
  is ever emitted, and nothing is ranked by a computed number.
- **Not a writer.** It writes nothing, connects to no database, runs no
  Power BI.

## See also

- `docs/capabilities/README.md` -- the fail-closed truthfulness contract and
  how this differs from `status`/`next`/`doctor`/`check`.
- `docs/capabilities/capabilities.yaml` -- the manifest this skill renders.

来源：skillsdirectory

能力

📘如何使用这个技能

评论0