ClawHub

ask-deep-wiki

v1.0.2

For questions about Light Protocol's SDK, smart contracts and Solana development, Claude Code features, or agent skills. AI-powered answers grounded in repos...

⭐ 0· 327·1 current·1 all-time
by@tilo-14
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/description (Light Protocol, Solana, Claude Code, agent skills) align with the runtime instructions: the SKILL.md explicitly calls DeepWiki MCP functions and repo paths for those projects. No unrelated binaries, credentials, or external services are requested.
✓
Instruction Scope
Instructions restrict activity to: (a) searching/reading files in the current repository (Glob/Grep/Read), (b) querying DeepWiki MCP for specific public repos, and (c) using limited WebFetch targets. The skill does not instruct the agent to read system config, environment secrets, or send data to arbitrary endpoints.
✓
Install Mechanism
No install spec and no code files — instruction-only skill. This is the lowest-risk install model and matches the described behavior.
✓
Credentials
Requires no environment variables, no credentials, and no config paths. The declared allowed tools (Read/Glob/Grep/WebFetch for two domains/mcp handlers) are proportionate to a documentation-retrieval skill.
✓
Persistence & Privilege
always:false and model invocation not disabled (normal defaults). The skill does not request persistent presence or modify other skills or system-wide settings.
Assessment
This skill appears coherent and does not request secrets or install code. Before installing: (1) confirm you trust the skill source — it will read files in whatever repository/context the agent is run against, so avoid running it in directories containing private keys or secrets; (2) note it queries DeepWiki MCP and GitHub repos (allowed WebFetch targets are limited), so ensure those endpoints are acceptable for your environment; (3) if you want extra assurance, review SKILL.md locally and run the skill in a sandboxed environment first.

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

latestvk978twmaekzgb1eyrtcqby8tk181te02

License

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

Comments