Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Deep Research (XiaoDing)
v1.0.1Deep Research Agent specializes in complex, multi-step research tasks that require planning, decomposition, and long-context reasoning across tools and files...
⭐ 0· 67·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The skill claims to integrate with a Search API and the File System and to persist cross-thread memory, but the registry metadata lists no required environment variables, credentials, or config paths. The rules/logic.md requires a CRAFTED_API_KEY for an external 'Crafted MCP server' which is not declared in the skill's requirements — an incoherent mapping between purpose and requested access.
Instruction Scope
Runtime instructions (rules/logic.md and SKILL.md) direct the agent to aggregate findings across files, tools, and conversations and to persist knowledge across sessions, and they explicitly require use of an external server. These steps imply reading local files and sending data externally, but there are no concrete, declared limits or endpoints and no explanation of what data is transmitted or stored.
Install Mechanism
This is an instruction-only skill with no install spec and no code files to execute; there is no installer or downloaded artifact visible in the package, which reduces install-time risk.
Credentials
Although the skill package metadata declares no required environment variables, the included logic explicitly demands a CRAFTED_API_KEY and directs the user to obtain it from we-crafted.com. Requesting an external API key that is not declared in the registry metadata (and that would likely grant remote access/storage) is disproportionate and unexplained.
Persistence & Privilege
The skill promises 'Cross-Thread Memory Persistence' and instructs the agent to 'Persist Knowledge' across conversations and to use the 'Crafted MCP server and your local environment.' That combination implies persistent storage of conversation data and possibly uploading local context to an external server — a significant privilege that is not transparently described in the registry metadata.
What to consider before installing
Before installing, ask the skill author to clarify and fix these inconsistencies: (1) Declare any required environment variables (e.g., CRAFTED_API_KEY) and explain why they are needed. (2) Specify exactly what data is persisted, where (local path vs external server), and retention/ownership policies. (3) Provide the API endpoints the agent will call and a privacy/security policy for we-crafted.com. (4) If the skill will read local files, ask for the exact paths or patterns and whether that data will be transmitted off-host. (5) Prefer installing only after code or concrete workflow is published (or run the skill in a sandboxed environment) and audit network traffic to verify no unintended data exfiltration. I have medium confidence because the package is instruction-only (no code to analyze) but the included instructions explicitly reference an undeclared external API key and cross-session persistence — clarifying those points could move this assessment to benign.Like a lobster shell, security has layers — review code before you run it.
latestvk97fz090mkwhj17rddpk3cm3xh83ftye
License
MIT-0
Free to use, modify, and redistribute. No attribution required.