Skip to main content

sage-wiki-integrate

Pipeline skill that wires sage-wiki into an existing project — detects language, installs the client, runs a smoke test, and reports.

跳到安装

来源信息

仓库
xoai/sage-wiki
最近来源活动
2026年8月1日 06:27
检测到的 SKILL.md 语言
英语
星标
608
分支
100

安装方式

默认使用会先检查来源的 Prompt;你也可以切换为直接命令,或下载本地副本。

检查来源文件

决定是否安装前,请先阅读 SKILL.md,以及 SkillsMP 当前展示的配套文件。

正在显示 SKILL.md

SKILL.md
来源说明 · 只读预览
name
sage-wiki-integrate
description
Pipeline skill that wires sage-wiki into an existing project — detects language, installs the client, runs a smoke test, and reports.
version
0.1.0
generated
skillgen — do not hand-edit
# sage-wiki-integrate Wires sage-wiki into an existing repository so the coding assistant can store and retrieve knowledge with provenance. Test-first — every step is verified before proceeding. ## Step 1: Detect language Check the project root for language markers: | Marker | Language | Integration path | |--------|----------|-----------------| | `pyproject.toml` | Python | `pip install sagewiki` (Python client, P4-3) | | `package.json` | TypeScript / JavaScript | `npm install sagewiki` (TypeScript client, P4-4) | | Neither | Any | MCP stdio config | ## Step 2: Choose integration path ### Python (`pyproject.toml` found) ```bash pip install sagewiki ``` ```python from sagewiki import Client client = Client(base_url="http://localhost:3333/v1") # If token-protected: # client = Client(base_url="http://localhost:3333/v1", token="s3cret") ``` ### TypeScript (`package.json` found) ```bash npm install sagewiki ``` ```typescript import { Client } from "sagewiki"; const client = new Client({ baseUrl: "http://localhost:3333/v1" }); // If token-protected: // const client = new Client({ baseUrl: "http://localhost:3333/v1", token: "s3cret" }); ``` ### Neither (MCP stdio) Add to the project's MCP config: ```json { "mcpServers": { "sage-wiki": { "command": "sage-wiki", "args": ["serve", "--transport", "stdio", "--project", "."] } } } ``` ## Step 3: Write a smoke test Store a fact, search for it, verify it comes back: **Python smoke test:** ```python # Store a fact via capture resp = client.capture(content="smoke-test: sage-wiki integration works at {now}") assert resp.status == "ok" # Search for it results = client.search(query="smoke-test sage-wiki integration") assert len(results) > 0, "smoke test failed: no results for stored fact" print("PASS: smoke test") ``` **TypeScript smoke test:** ```typescript const resp = await client.capture({ content: "smoke-test: sage-wiki integration works" }); const results = await client.search({ query: "smoke-test sage-wiki integration" }); console.assert(results.length > 0, "smoke test failed"); console.log("PASS: smoke test"); ``` **MCP smoke test** (call `wiki_capture` then `wiki_search` via your MCP client): ``` Tool: wiki_capture Args: { content: "smoke-test: sage-wiki integration works at {timestamp}" } Expected: confirmation text, no error Tool: wiki_search Args: { query: "smoke-test sage-wiki integration" } Expected: at least one result, no error ``` ## Step 4: Run the smoke test Execute the chosen smoke test. If it fails: - Check the server is running (`sage-wiki serve --ui`) - Verify the token is correct (if auth is enabled) - Check the server logs for errors ## Step 5: Report After a successful smoke test, confirm: - sage-wiki is connected and reachable - The chosen integration path works - Store → search → verify cycle completes Then refer to the `sage-wiki` reference skill for the full tool catalog, error codes, and opt-in features. > **Pre-1.0 notice:** Every surface here is experimental. Tool semantics, argument names, and REST routes may change between versions. Pin the version you depend on. Generated by tools/skillgen — reference skill covers 19 tools.
在 GitHub 查看