Zhihu

Use this skill for Zhihu question research, answer mining, 专栏文章采集, and topic discovery. It fits analyst workflows where the user starts from a keyword, question ID, article ID, or column ID and wants structured Zhihu data rather than a high-level summary.

When To Use It

• The user wants Zhihu keyword search, question answer extraction, or column article research.
• The task is about a specific Zhihu question, answer feed, 专栏文章, or column page.
• The user can provide keyword, questionId, id, columnId, or pagination values such as cursor and offset.
• The user needs Zhihu-native titles, content, authors, ranking context, or article metadata sourced directly from the API.

Representative Operations

• searchV1: Keyword Search — Search Zhihu by keyword for topic discovery, source gathering, and question scouting.
• getAnswerListV1: Answer List — Pull the answer feed for a Zhihu question to inspect authors, content, and interaction signals.
• getColumnArticleDetailV1: Column Article Details — Read a single Zhihu column article in detail for archiving and content analysis.
• getColumnArticleListV1: Column Article List — Track the article feed of a Zhihu column for monitoring and collection work.

Request Pattern

• 4 read-only GET operations are available in this skill.
• Common inputs are keyword, questionId, columnId, id, cursor, offset, and order.
• The workflow splits cleanly into search, question-answer feeds, single article detail, and column article listing.
• No operation in this skill requires a request body.

How To Work

1. Read generated/operations.md before choosing an endpoint.
2. Start with one of these operations when it matches the user's request: searchV1, getAnswerListV1, getColumnArticleDetailV1, getColumnArticleListV1.
3. Pick the smallest matching operation instead of guessing.
4. Ask the user for any missing required parameter. Do not invent values.
5. Call the helper with:

node {baseDir}/bin/run.mjs --operation "<operation-id>" --token "$JUST_ONE_API_TOKEN" --params-json '{"key":"value"}'

Environment

• Required: JUST_ONE_API_TOKEN
• This skill uses JUST_ONE_API_TOKEN only for authenticated Just One API requests.
• Keep JUST_ONE_API_TOKEN private. Do not paste it into chat messages, screenshots, or logs.
• Get a token from Just One API Dashboard.
• Authentication details: Just One API Usage Guide.

Output Rules

• Lead with the concrete Zhihu result: matched topic, answer feed insight, article detail, or column update.
• For keyword search, restate the keyword and offset before summarizing the most relevant matches.
• For answer-list requests, surface the question scope, top authors, engagement cues, and any ordering rule before raw JSON.
• For column-article requests, extract title, author, publication context, and the most important content signals before raw JSON.
• If the backend errors, include the backend payload and the exact operation ID.