ClawHub

Feishu Notify

v1.0.1

This skill should be used when users need to send notifications to Feishu (Lark) via webhook. It supports multiple message types including text, rich text, i...

⭐ 1· 112·1 current·1 all-time
by@alex098929
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name and description (send messages to Feishu via webhook) match the included files and runtime behavior. The skill only needs webhook URLs stored in a user config file and uses templates and a small Python script to POST messages to Feishu webhooks.
ℹ
Instruction Scope
SKILL.md instructs the agent to read a single config file in the user's home (~/.openclaw/skills/feishu-notify/config.json) and to load templates from the skill's templates/ directory — this matches the script. One implementation issue: load_template uses a string startswith() check to ensure template paths are in the templates directory; that can be bypassed by paths whose absolute path begins with the templates path string (e.g., '/.../templates_malicious/...') or via certain symlink scenarios. Otherwise the instructions do not ask for unrelated files or credentials.
✓
Install Mechanism
No install spec (instruction-only plus a small script). No external downloads or package installs are performed by the skill bundle itself.
✓
Credentials
The skill requests no environment variables or platform credentials. The only secret required is the Feishu webhook URL stored by the user in a local config file — this is proportional to the stated purpose.
ℹ
Persistence & Privilege
The skill is not forced-always and does not request special privileges, but SKILL.md explicitly states the agent can call this skill autonomously. That means an autonomous agent could send messages (to whatever webhooks are configured) without explicit user confirmation — expected for messaging skills but worth user awareness.
Assessment
This skill appears to do what it says: send messages to Feishu webhooks using a local config file and templates. Before installing: - Keep your webhook URLs secret and stored only in ~/.openclaw/skills/feishu-notify/config.json with restrictive permissions (chmod 600). The webhook URL itself is the authentication token for posting. - Be aware that the agent may invoke this skill autonomously and send messages without an explicit prompt if the agent chooses to; if you need manual confirmation, disable autonomous invocation or only configure non-critical webhooks. - The template loader has a weak directory check (uses startswith); consider reviewing or hardening the code (use os.path.commonpath or compare path components / add trailing slash) if you worry about template-path trickery or symlink-based escapes. - Test using a disposable webhook or a non-production Feishu group before enabling in production. If you want a higher security posture, review the script locally (send_message.py) and optionally modify it to: validate template paths via os.path.commonpath, avoid accepting arbitrary template path input from untrusted prompts, and log less/avoid printing secrets. Overall the skill is internally consistent and there are no unexplained external endpoints or extra credentials requested.

Like a lobster shell, security has layers — review code before you run it.

latestvk979sw9z2wrq9hkwz88c4k8b4d836z4k

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Comments