Intent-routed web search via Exa MCP. 8 modes (quick/deep/code/docs/debug/news/compare/research).
Mirko Bozzetto's curated skills for AI coding agents.
Skills are markdown-based, portable across providers (Claude Code, Codex, Cursor, etc.). Native plugin install is wired for Claude Code; other providers can load the skill files directly.
First plugin: /websearch - a power web search skill that wraps Exa MCP under a single command with 8 intent-routed modes. Replaces native WebSearch and WebFetch with semantic search, structured deep research, code lookup, docs crawling, and inline-cited reports.
# 1. (Optional) Sign up at https://dashboard.exa.ai and copy an API key
# Free anonymous tier works without a key, rate-limited.
# Add a key only when you hit rate limits.
# 2. Add the Exa MCP server (hosted, recommended)
claude mcp add --transport http exa https://mcp.exa.ai/mcp
# 3. Verify connection
claude mcp list
# 4. Inside Claude Code:
/plugin marketplace add mirkobozzetto/arsenal
/plugin install websearch@arsenal
# 5. Try it
/websearch what is gRPC
Full setup walkthrough, mode reference, troubleshooting → plugins/websearch/README.md.
Skills here are plain markdown. To use them outside Claude Code:
plugins/websearch/skills/websearch/ into your agent's skill/prompt directory.The skill body assumes the 4 Exa MCP tools (web_search_exa, get_code_context_exa, crawling_exa, web_search_advanced_exa). Any agent that can call those will work.
/websearch doesOne slash command, 8 modes selected by flag:
| Mode | Flag | Use case |
|---|---|---|
| Quick | (none) | Factual lookup, one-shot |
| Deep research | --deep | Multi-pass with gap analysis |
| Code | --code | API snippets, library usage |
| Docs | --docs <lib> | Crawl official docs |
| Debug | --debug <err> | Paste error, find fix |
| News | --news | Recent news, category filtered |
| Compare | --compare A vs B | Side-by-side tech choice |
| Research | --research | Academic papers |
| Similar | --similar <url> | Find similar pages |
Filters (combinable): --after, --before, --domain, --exclude, --fresh, --locale, -n.
Output: --save <file>, --json, --full.
Help: --info (full tutorial inside the skill).
| Plugin | Version | Description |
|---|---|---|
| websearch | 1.0.0 | Intent-routed web search via Exa MCP (8 modes) |
More plugins may land here over time. Marketplace name stays arsenal.
mcp__exa__* tools available).https://mcp.exa.ai/mcp works anonymously with a free, rate-limited tier. Add a key via header Authorization: Bearer KEY or x-api-key: KEY to bypass rate limits.Get a free key at dashboard.exa.ai/api-keys. See docs.exa.ai/reference/exa-mcp for Exa MCP details.
/plugin update works out of the box.Open an issue or PR at github.com/mirkobozzetto/arsenal. One concern per PR. Bump semver on any change shipping to users.
MIT - see LICENSE.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub mirkobozzetto/arsenal --plugin websearchConcentrated Claude. Zero fluff, max power per token.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Build and maintain an LLM-curated personal knowledge base in your project — Andrej Karpathy's LLM Wiki pattern, designed to scale to thousands of pages without becoming a context bottleneck. Now with an optional compiled graph layer for typed, provenance-backed relationships.
Connect to Atlassian products including Jira and Confluence. Search and create issues, access documentation, manage sprints, and integrate your development workflow with Atlassian's collaboration tools.
AI-powered wiki generator for code repositories. Generates comprehensive, Mermaid-rich documentation with dark-mode VitePress sites, onboarding guides, deep research, and source citations. Inspired by OpenDeepWiki and deepwiki-open.
20 modular skills for idiomatic Go — each under 225 lines, backed by 48 reference files, 8 automation scripts (all with --json, --limit, --force), and 4 asset templates. Covers error handling, naming, testing, concurrency, interfaces, generics, documentation, logging, performance, and more. Activates automatically with progressive disclosure and conditional cross-references.
Claude + Obsidian knowledge companion. Sets up a persistent, compounding wiki vault (Karpathy's LLM Wiki pattern). v1.7 "Compound Vault" + v1.8 methodology modes close 5 of 5 priority gaps from the May 2026 compass artifact. Ships: substrate alignment with kepano/obsidian-skills, default Obsidian CLI transport, hybrid retrieval (contextual prefix + BM25 + cosine rerank per Anthropic's Sept 2024 research), per-file advisory locking for multi-writer safety, pre-commit verifier agent, AND methodology modes (LYT / PARA / Zettelkasten / Generic) for first-class organizational support no other Claude+Obsidian competitor offers. v1.7.x audit closure: every BLOCKER + HIGH + MEDIUM + LOW finding from the v1.7.0 audit is CLOSED or DEFERRED-with-rationale. Optional DragonScale Memory extension (log folds, deterministic addresses, semantic tiling lint, boundary-first autoresearch).