openclaw-dingtalk-channel
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions for the agent or user to download and install a plugin from a third-party source at
https://github.com/soimy/openclaw-channel-dingtalk.gitand the NPM package@soimy/dingtalk. While this is the intended purpose of the skill, it involves the execution of external code from a non-pre-approved source. - [INDIRECT_PROMPT_INJECTION]: The skill implements a communication channel that ingests untrusted data from DingTalk and processes it within the agent's workflow.
- Ingestion points: External messages are ingested via the
processInboundfunction described inSKILL.md. - Boundary markers: The provided implementation examples do not utilize specific delimiters or instructions to the agent to ignore potentially malicious embedded content within the DingTalk messages.
- Capability inventory: The skill allows the agent to send messages, interact with DingTalk interactive cards, and process user-provided files and voice notes.
- Sanitization: The documented logic performs basic text cleanup (trimming and prefix stripping) but lacks rigorous sanitization or validation of the external input content.
Audit Metadata