lark-wiki
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from external wiki URLs and node content (Ingestion points: SKILL.md, lark-wiki-node-get.md). The skill mitigates risks by requiring explicit user selection of resolved candidates and the --yes confirmation flag for modifications (Boundary markers: present in deletion and space matching workflows). The skill possesses significant capabilities including node deletion, space deletion, and member management (Capability inventory: lark-cli wiki +node-delete, +delete-space, +member-add). It validates input by strictly parsing internal IDs from token resolution results before execution (Sanitization: space_id resolution logic in lark-wiki-delete-space.md).
- [COMMAND_EXECUTION]: The skill utilizes lark-cli to perform wiki operations. Safety is maintained by differentiating between user and bot identities and requiring confirmation flags for all high-risk write operations.
- [SAFE]: The skill architecture and behavior are consistent with the larksuite vendor's requirements, following best practices for resource management and identity scoping.
Audit Metadata