Skill hub for the solana-ai-kit Claude Code plugin. Routes the bundled go-to-market skills and the opt-in add-on catalog, and points to upstream marketplaces or the install.sh full install for protocol, security and ecosystem depth. Use it when a kit agent, command or skill links to a missing file in an ext skill pack.
107 stars
0 votes
0 copies
0 views
Added October 2, 2026
ai-agentsgobashtestinggitsecurity
Works with
claude code
cli
mcp
Security analysis
C71/100
criticalPipes output to a shell interpreter
mediumUses curl or wget to download content
criticalDownloads and executes remote scripts — classic supply chain attack
Installs into .claude/skills of the current project.
Are you the author of Solana Ai Kit?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/solanabr-solana-ai-kit-0e170827)
---
name: solana-ai-kit
description: Skill hub for the solana-ai-kit Claude Code plugin. Routes the bundled go-to-market skills and the opt-in add-on catalog, and points to upstream marketplaces or the install.sh full install for protocol, security and ecosystem depth. Use it when a kit agent, command or skill links to a missing file in an ext skill pack.
user-invocable: true
---
# Solana AI Kit plugin skill hub
This is the plugin variant of the kit's skill hub. It ships only the skills that travel cleanly in a plugin: the go-to-market skills, the Token Extensions skill and the opt-in add-on catalog. The protocol, security and ecosystem skills, the project CLAUDE.md with the program-code house rules, and the curated permissions and sandbox policy come with the full install (see "Getting more depth").
When sources overlap: a protocol's official skill wins for its own SDK (Jupiter, Metaplex, Helius); the Solana Foundation `solana-dev` skill wins for general Solana work (Anchor, Pinocchio, testing, clients); community skills such as sendai fill gaps only. In plugin form, add the relevant upstream marketplace first.
## Bundled skills
These load when the plugin is enabled. Commands and skills are namespaced under `solana-ai-kit:` (for example `/solana-ai-kit:deploy`).
- [skill-registry.json](../skill-registry.json): catalog of opt-in add-on skills, plugins and MCPs that are not bundled. Search it by domain or tag and run an entry's install command only after the user confirms and `safe-ai-skill add skill|mcp <source>` returns `proceed: true`. Entries with a `tier` are the full install's pinned skill packs; their `source` is the upstream repo.
The kit's own [token-extensions](../token-extensions/SKILL.md) covers Token-2022: which extensions to use and how they combine, then per-extension CLI, Kit and Anchor setup. Its links to the Solana Foundation solana-dev skill need the `install.sh` full install.
## Security firewall (core)
This plugin depends on [safe-ai-skill](https://github.com/solanabr/safe-ai-skill) (`safe-ai-skill@stbr`), which installs with it. Its hooks gate mainnet, value-moving and authority actions and secret reads, and pin installed skills at session start. Its ask or deny is the user's policy, so don't retry the action another way. Its CLI is on PATH while it is enabled: `safe-ai-skill status`, `safe-ai-skill verify check <dir>`.
## Getting more depth
Plugins cannot carry git submodules, so the kit's external skill packs (the `ext` submodules) are not bundled here. Two ways to get them:
### Option A: add the upstream marketplaces
| Domain | Add the marketplace | Then install |
|--------|---------------------|--------------|
| DeFi protocols, infra, data (Jupiter, Raydium, Kamino, perps, oracles, cross-chain) | `/plugin marketplace add sendaifun/skills` | the protocol plugins you need |
| Security audits of programs and the code around them | nothing to add: a full install carries `auditor-skill` as a core pack (20 checklists, 1,424 items, 138 known vectors) | — |
| Infrastructure (Workers, Agents SDK, MCP servers) | `/plugin marketplace add cloudflare/skills` | `cloudflare` |
For Jupiter, Metaplex, Helius, MagicBlock and Alchemy the official skill repos are the primary sources; `skill-registry.json` lists their current upstream locations. Route to a marketplace and install the plugin rather than pointing at an upstream repo's `SKILL.md`.
### Option B: full install (recommended for project teams)
Run the installer in your project to get what the plugin can't carry: the core external skill packs (extensions install on demand with `/add-skill`), the project CLAUDE.md, and the curated permissions and sandbox policy.
```bash
curl -fsSL https://raw.githubusercontent.com/solanabr/ai-kit/main/install.sh | bash
```
The project README ("External Skill Submodules" and "Install as a Claude Code plugin") explains when to pick the plugin or the full install; they are complementary. If both are active in one project, `/solana-ai-kit:doctor` flags the duplicate commands, hooks and MCP servers.
## When a link into an `ext` pack is missing
The plugin's agents, commands and bundled skills are the same files the full install uses. Many of their links go through `skills/ext` into the external skill packs, and some lines add "install first: `bash .claude/bin/skills.sh add <id>`". A plugin install has neither the packs nor `.claude/bin`, so the link is dead and that command cannot run. Use a fallback and keep going:
- General Solana work (the `solana-dev` pack: Anchor, Pinocchio, `@solana/kit`, testing, security): ask the bundled solana-dev MCP.
- The exact file: the kit's site serves the packs under `https://aikit.superteam.codes/.claude/skills/ext`; append the part of the link that follows `ext` (for a folder link, its `SKILL.md`). If the site doesn't have it, use the pack's own repository, the `source` of its `skill-registry.json` entry (the folder after `ext` is the pack id); paths there can differ from the commit the kit pins.
- Tell the user once that the full install (`install.sh`) puts the packs in the project, where these links and `skills.sh` work.
## Task routing
| User asks about | Skill |
|-----------------|-------|
| An add-on skill, plugin or MCP that isn't bundled | [skill-registry.json](../skill-registry.json) |
| A safe-ai-skill ask or deny, skill or MCP supply-chain checks | Security firewall (core) above |
| Protocol SDK depth, security audits, infra | Option A marketplaces or the Option B full install |
| Token-2022 extensions: fees, hooks, metadata, pausable, confidential | [token-extensions](../token-extensions/SKILL.md) |