agent-search 0.4.4

A single Rust binary that aggregates 11 search providers into one unified search interface. Designed from day one for AI agents — structured JSON output, semantic exit codes, machine-readable capabilities, and auto-JSON when piped.

Works with OpenClaw, Claude Code, Codex CLI, Gemini CLI, or any agent framework that can shell out to a command.

search "CRISPR gene therapy breakthroughs"

That's it. Auto-detects your intent, fans out to the right providers in parallel, deduplicates results, and returns them in under 2 seconds.

Why

Every search API is good at something different. Brave has its own 35-billion page index. Serper gives you raw Google results plus Scholar, Patents, and Places. Exa does neural/semantic search and finds LinkedIn profiles. Perplexity gives AI-synthesized answers with citations using Sonar Pro. Jina reads any URL into clean markdown. Firecrawl renders JavaScript-heavy pages. xAI searches X/Twitter via Grok.

search routes your query to the right combination automatically — or lets you pick exactly which providers to use.

Install

Cargo (recommended):

cargo install agent-search

One-liner (macOS / Linux):

curl -fsSL https://raw.githubusercontent.com/199-biotechnologies/search-cli/master/install.sh | sh

Homebrew:

brew tap 199-biotechnologies/tap
brew install search-cli

From source:

cargo install --git https://github.com/199-biotechnologies/search-cli

Binary size: ~6MB. Startup: ~2ms. Memory: ~5MB. No Python, no Node, no Docker.

Quick Start

# 1. Set your API keys (any combination works — even just one)
search config set keys.brave YOUR_BRAVE_KEY
search config set keys.serper YOUR_SERPER_KEY
search config set keys.exa YOUR_EXA_KEY
search config set keys.jina YOUR_JINA_KEY
search config set keys.firecrawl YOUR_FIRECRAWL_KEY
search config set keys.tavily YOUR_TAVILY_KEY
search config set keys.serpapi YOUR_SERPAPI_KEY
search config set keys.perplexity YOUR_PERPLEXITY_KEY
search config set keys.browserless YOUR_BROWSERLESS_KEY
search config set keys.xai YOUR_XAI_KEY

# Or use environment variables
export SEARCH_KEYS_BRAVE=YOUR_KEY
export SEARCH_KEYS_EXA=YOUR_KEY
export SEARCH_KEYS_XAI=YOUR_KEY

# 2. Search
search "your query here"

Usage

# Auto-detect mode (recommended — just type what you want)
search "quantum computing advances"
search "who is the CEO of Anthropic"
search "latest AI news today"
search "CRISPR research papers"
search "trending on twitter AI"

# Force a specific mode
search search -q "transformer architectures" -m academic
search search -q "Sam Altman" -m people
search search -q "AI startups 2026" -m news
search search -q "BRCA1 gene patent" -m patents
search search -q "coffee shops near me" -m places
search search -q "https://example.com" -m extract
search search -q "what are people saying about Rust" -m social

# Pick specific providers
search search -q "machine learning" -p exa              # Exa only
search search -q "rust programming" -p brave,serper      # Brave + Serper
search search -q "trending AI" -p xai                    # xAI X/Twitter only
search search -q "https://arxiv.org/..." -p jina         # Jina reader only

# Control output
search "query" --json                  # Force JSON output
search "query" --json | jq '.results[].url'   # Pipe to jq
search "query" -c 20                   # Get 20 results
search "query" 2>/dev/null             # Suppress diagnostics

Auto-JSON: Output is automatically JSON when piped to another program. Human-readable tables when you're in a terminal.

Modes

Providers

Agent Integration

Built for AI agents from day one. Every command supports --json and structured error codes.

# Discover capabilities programmatically
search agent-info

# Structured JSON with metadata
search "query" --json
# {
#   "version": "1",
#   "status": "success",
#   "query": "...",
#   "mode": "general",
#   "results": [...],
#   "metadata": {
#     "elapsed_ms": 1542,
#     "result_count": 10,
#     "providers_queried": ["brave", "serper", "exa"],
#     "providers_failed": []
#   }
# }

# Errors are also structured JSON
search "query" --json 2>&1
# {
#   "status": "error",
#   "error": {
#     "code": "no_providers",
#     "message": "No providers configured for mode 'general'",
#     "suggestion": "Configure at least one provider API key"
#   }
# }

Exit codes are semantic:

Configuration

Config file lives at ~/.config/search/config.toml (Linux) or ~/Library/Application Support/search/config.toml (macOS).

search config show       # View current config (keys masked)
search config check      # Health check all providers
search config set K V    # Set a value

Environment variables override the config file. Prefix: SEARCH_KEYS_:

export SEARCH_KEYS_BRAVE=your-key
export SEARCH_KEYS_SERPER=your-key
export SEARCH_KEYS_EXA=your-key
export SEARCH_KEYS_JINA=your-key
export SEARCH_KEYS_FIRECRAWL=your-key
export SEARCH_KEYS_TAVILY=your-key
export SEARCH_KEYS_SERPAPI=your-key
export SEARCH_KEYS_PERPLEXITY=your-key
export SEARCH_KEYS_BROWSERLESS=your-key
export SEARCH_KEYS_XAI=your-key

How It Works

1. Parse — Clap parses your query, mode, provider filter, and output preferences
2. Classify — If mode is auto, regex-based intent classifier picks the right mode
3. Route — Mode determines which providers to query (or you override with -p)
4. Fan out — tokio::JoinSet fires all providers in parallel with per-provider timeouts
5. Collect — Results stream in as providers respond (no waiting for the slowest)
6. Dedup — URL normalization removes duplicates across providers
7. Log — Every search is logged to ~/Library/Application Support/search/logs/ (JSONL)
8. Render — JSON envelope or colored terminal table, auto-detected from context

Updating

search update             # Self-update from GitHub releases
search update --check     # Check without installing

Building from Source

git clone https://github.com/199-biotechnologies/search-cli
cd search-cli
cargo build --release
# Binary at target/release/search

License

MIT

Created by Boris Djordjevic at 199 Biotechnologies.