Parsa-29/loop-config / skills/bx
bx
FlaggedUSE FOR web search, research, RAG, grounding, browse, find, lookups, fact-checking, documentation, agentic AI. All-in-one, optimized for AI agents. Pre-extracted, token-budgeted web content, deep research, news, images, videos, places, custom ranking
bx — Brave Search CLI
Use bx for all web searches. Run bx "query" as the default — it returns pre-extracted, token-budgeted web content ready for use. Use bx answers for synthesized explanations or bx web for traditional search results. If bx is not found, install it: curl -fsSL https://raw.githubusercontent.com/brave/brave-search-cli/main/scripts/install.sh | sh
Requires: Brave Search API Key +
bxbinaryPlans: Different subcommands may require different plans. See https://api-dashboard.search.brave.com/app/subscriptions/subscribe
Quick Start
# Install (macOS/Linux)
curl -fsSL https://raw.githubusercontent.com/brave/brave-search-cli/main/scripts/install.sh | sh
# Configure API key (get one at https://api-dashboard.search.brave.com)
bx config set-key # interactive (avoids shell history)
# or: export BRAVE_SEARCH_API_KEY=YOUR_KEY
# Search (default = bx context "query")
bx "your search query"
When to Use Which Command
| Your need | Command | Why |
|---|---|---|
| Look up docs, errors, code patterns | bx "query" | Pre-extracted text, token-budgeted (default) |
| Get a synthesized explanation | bx answers "query" | AI-generated, cites sources, streams |
| Deep research on complex topics | bx answers "query" --enable-research | Multi-search iterative research |
| Traditional search results | bx web "query" | All result types (web, news, discussions, etc.) |
| Find discussions/forums | bx web "query" --result-filter discussions | Forums often have solutions |
| Latest news / recent events | bx news "query" --freshness pd | Fresh info beyond training data |
| Find images | bx images "query" | Up to 200 results |
| Find videos | bx videos "query" | Duration, views, creator |
| Local businesses / places | bx places "coffee" --location "San Francisco" | 200M+ POIs |
| Boost/filter specific domains | bx "query" --include-site docs.rs | Or use --goggles for full control |
Commands
| Command | Description | Output path |
|---|---|---|
context | Default. RAG/LLM grounding — pre-extracted web content | .grounding.generic[] -> {url, title, snippets[]} |
answers | AI answers — OpenAI-compatible, streaming by default | .choices[0].delta.content (stream) |
web | Web search — all result types | .web.results[], .news.results[], etc. |
news | News articles with freshness filters | .results[] -> {title, url, age} |
images | Image search (up to 200 results) | .results[] -> {title, url, thumbnail.src} |
videos | Video search with duration/views | .results[] -> {title, url, video.duration} |
places | Local place/POI search (200M+ POIs) | .results[] -> {title, postal_address} |
suggest | Autocomplete/query suggestions | .results[] -> {query} |
spellcheck | Spell-check a query | .results[0].query |
config | Manage API key and settings | set-key, show-key, path, show |
Response Shapes
bx "query" (context — default, recommended)
{
"grounding": {
"generic": [
{ "url": "...", "title": "...", "snippets": ["extracted content...", "..."] }
]
},
"sources": {
"https://example.com": { "title": "...", "hostname": "...", "age": ["...", "2025-01-15", "392 days ago"] }
}
}
bx answers "query" --no-stream (single JSON response)
{"choices": [{"message": {"content": "Full answer text..."}}]}
bx answers "query" (streaming — default, one JSON chunk per line)
{"choices": [{"delta": {"content": "chunk"}}]}
bx web "query" (full search results)
{
"web": { "results": [{"title": "...", "url": "...", "description": "..."}] },
"news": { "results": [...] },
"videos": { "results": [...] },
"discussions": { "results": [...] }
}
Token Budget Control
Control output size for context (the default command):
| Flag | Short alias | Default | Description |
|---|---|---|---|
--maximum-number-of-tokens | --max-tokens | 8192 | Approximate total tokens (1024-32768) |
--maximum-number-of-tokens-per-url | --max-tokens-per-url | 4096 | Max tokens per URL (512-8192) |
--maximum-number-of-urls | --max-urls | 20 | Max URLs in response (1-50) |
--maximum-number-of-snippets | --max-snippets | 50 | Max snippets across all URLs |
--maximum-number-of-snippets-per-url | --max-snippets-per-url | — | Max snippets per URL |
--context-threshold-mode | --threshold | balanced | Relevance: strict, balanced, lenient |
bx "topic" --max-tokens 4096 --max-tokens-per-url 1024 --max-urls 5 --threshold strict
Goggles — Custom Ranking
Goggles let you control which sources appear in results. Boost official docs, suppress SEO spam, or build focused search scopes. No other search tool offers this. Supported on context, web, and news.
Domain Shortcuts
# Allowlist — only results from these domains
bx "rust axum" --include-site docs.rs --include-site github.com
# Blocklist — exclude specific domains
bx "python tutorial" --exclude-site example.com
--include-site, --exclude-site, and --goggles are mutually exclusive.
Inline Rules
# Boost official docs, demote blog posts
bx "axum middleware tower" \
--goggles '$boost=5,site=docs.rs
$boost=3,site=github.com
/docs/$boost=5
/blog/$downrank=3' --max-tokens 4096
# Allowlist mode — only include matched sites
bx "Python asyncio patterns" \
--goggles '$discard
$boost,site=docs.python.org
$boost,site=peps.python.org'
DSL Quick Reference
| Rule | Effect | Example |
|---|---|---|
$boost=N,site=DOMAIN | Promote domain (N=1-10) | $boost=3,site=docs.rs |
$downrank=N,site=DOMAIN | Demote domain (N=1-10) | $downrank=5,site=example.com |
$discard,site=DOMAIN | Remove domain entirely | $discard,site=example.com |
/path/$boost=N | Boost matching URL paths | /docs/$boost=5 |
Generic $discard | Allowlist mode — discard unmatched | $discard (as first rule) |
Separate rules with newlines. Full DSL: goggles-quickstart.
Piping Rules via Stdin
echo '$boost=5,site=docs.rs
$boost=5,site=crates.io
$boost=3,site=github.com' | bx "axum middleware" --goggles @- --max-tokens 4096
Use @/path/to/file to reuse a goggle across queries.
Agent Workflow Examples
Debugging an error:
bx "Python TypeError cannot unpack non-iterable NoneType" --max-tokens 4096
Corrective RAG loop:
# 1. Broad search
bx "axum middleware authentication" --max-tokens 4096
# 2. Too general? Narrow with strict threshold
bx "axum middleware tower layer authentication example" --threshold strict --max-tokens 4096
# 3. Still need synthesis? Ask for an answer
bx answers "how to implement JWT auth middleware in axum" --enable-research
Checking for breaking changes before upgrading:
bx "Next.js 15 breaking changes migration guide" --max-tokens 8192
bx news "Next.js 15 release" --freshness pm
Non-streaming answers for programmatic use:
bx answers "compare SQLx and Diesel for Rust" --no-stream
Answers via stdin (OpenAI-compatible JSON body):
echo '{"messages":[{"role":"user","content":"what are the OWASP top 10 vulnerabilities for web APIs"}]}' | bx answers -
Exit Codes
| Code | Meaning | Agent action |
|---|---|---|
| 0 | Success | Process results |
| 1 | Client error (bad request) | Fix query/parameters |
| 2 | Usage error (bad flags) | Fix CLI arguments |
| 3 | Auth/permission error (401/403) | Check API key: bx config show-key |
| 4 | Rate limited (429) | Retry after delay |
| 5 | Server/network error | Retry with backoff |
Use Cases
- AI agents / coding assistants: One-call web search with token-budgeted, RAG-ready content — replaces search + scrape + extract
- Fact-checking: Verify claims against current web content with
bx "query" --threshold strict - Documentation lookup: Search official docs with
--include-siteor Goggles domain boosting - Research: Deep multi-source research with
bx answers "topic" --enable-research - Debugging: Search for error messages and stack traces directly
- News monitoring: Track topics with
bx news "query" --freshness pd - Local search: Find businesses and places with
bx places "query" --location "city"
Notes
- All output is JSON to stdout; errors go to stderr
- Global flags:
--api-key KEY,--timeout SECS(default 30),--extra KEY=VALUE(repeatable, adds arbitrary API parameters) - Location awareness:
contextandwebsupport--lat,--long,--city,--state,--state-name,--loc-country,--postal-code,--timezone - Research mode:
bx answers --enable-researchcan take up to 5 minutes; set client timeout accordingly - Help:
bx --helpfor all commands;bx <command> --helpfor per-command flags
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization. | Stars: 465k | Updated | Claude CodeCodex | Verified | Install |
| academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file). | Stars: 196k | Updated | Claude CodeCodex | Unverified | Install |
| doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |