Zhihu
ReviewAudited by ClawScan on May 10, 2026.
Overview
This Zhihu integration appears purpose-aligned, but it gives an agent credentialed ability to post, like, comment, and delete Zhihu content, so it should be used with supervision.
Install this only if you are comfortable letting an agent use your Zhihu API credentials to perform public/account-visible actions. Supervise publish, comment, reaction, and delete requests, verify target IDs and text before execution, and protect the configured app key and app secret.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
If invoked incorrectly, the agent could publish unwanted posts, like or unlike content, create comments, or delete comments through the user's Zhihu API access.
The skill explicitly grants tools that can change public or account-visible Zhihu content. This is expected for the stated purpose, but it is high-impact enough that users should supervise use.
Publishing pins ... Liking/unliking pins and comments ... Creating comments ... Deleting comments
Use this skill only when you want the agent to act on Zhihu. Review post/comment text and target IDs before allowing publish, reaction, or delete operations.
Anyone or any agent process with access to these environment variables can use the configured Zhihu API authority within the API's permissions.
The tool uses Zhihu API credentials from environment variables to authenticate requests. This is expected for the integration, but those credentials authorize account actions.
self.app_key = os.environ.get("ZHIHU_APP_KEY", "")
self.app_secret = os.environ.get("ZHIHU_APP_SECRET", "")Use least-privilege Zhihu credentials if available, keep the app secret private, and rotate the credentials if they are exposed.
A user may need to prepare the Python environment manually, and dependency versions are not pinned by the skill artifacts.
The README documents a manual, unpinned dependency installation, while no install spec is provided. This is not suspicious by itself, but users should install dependencies from trusted sources.
requests 库：`pip install requests`
Install dependencies in a trusted virtual environment and consider pinning or reviewing the requests package version used.