Deep research powered by AIresearchOS. Submit, track, and retrieve research with clarifying questions. Supports API key auth and x402 USDC payments.
数据来源:ClawHub。 在 ClawSkills 查看
选择你使用的 Agent
方法一:命令行安装(推荐)
推荐(无需提前安装 clawhub)
npx clawhub@latest --dir ~/.claude/skills install openclaw-airesearchos或使用 clawhub CLI(需提前安装)
clawhub --dir ~/.claude/skills install openclaw-airesearchos⚠️ 需要 Node.js 18+,没有 Node?请使用下方方法二直接下载 ZIP。 安装 Node.js →
方法二:手动下载安装(无需 Node)
下载 ZIP,解压后将文件夹放到以下路径,重启 Agent 即可:
安装路径
~/.claude/skills/openclaw-airesearchos/💡解压后将文件夹放到上方路径,重启 Agent 即可生效
--- name: airesearchos description: "Deep research powered by AIresearchOS. Submit, track, and retrieve research with clarifying questions. Supports API key auth and x402 USDC payments." homepage: https://airesearchos.com metadata: { "openclaw": { "emoji": "🔬", "requires": { "bins": ["curl"] }, "primaryEnv": "AIRESEARCHOS_API_KEY", "homepage": "https://airesearchos.com", }, } ---
Deep research API — submit queries, track progress, retrieve full reports with sources.
Before any research operation, determine the auth method:
# Check API key first (preferred)
if [ -n "$AIRESEARCHOS_API_KEY" ]; then
echo "api_key"
# Then check x402 wallet
elif [ -n "$AIRESEARCHOS_WALLET_KEY" ]; then
echo "x402"
else
echo "unconfigured"
fi
/api/v1/ endpoints with Authorization: Bearer $AIRESEARCHOS_API_KEY/api/x402/ endpoints via {baseDir}/scripts/x402-request.mjsPresent this to the user:
> To use AIresearchOS, you need to set up authentication. > > Option 1: API Key (recommended for regular use) > - Sign up at https://airesearchos.com > - Go to Dashboard → Settings → Generate API Key > - Add to ~/.openclaw/openclaw.json: > skills.entries.airesearchos.apiKey = "aro_sk_..." > - Start a new OpenClaw session > > Option 2: x402 Pay-Per-Request (no account needed) > - Pay with USDC stablecoins per request > - Scan: $0.50 | Due Diligence: $1.50 | Mission Critical: $5.00 > - Run: cat {baseDir}/SETUP.md for full x402 setup guide > > Which would you prefer?
| Mode | API Key Credits | x402 Cost | Depth/Breadth | Sources | Best For | |------|----------------|-----------|---------------|---------|----------| | scan | 10 credits | $0.50 USDC | 2/2 | 10-20 | Quick validation | | dueDiligence | 25 credits | $1.50 USDC | 3/3 | 50-100 | Decision-grade analysis | | missionCritical | 100 credits | $5.00 USDC | 5/5 | 150-300+ | Exhaustive coverage |
Report lengths: concise, standard (default), extended.
Confirm with user before: Mission Critical (100 credits / $5.00), any x402 payment.
curl -s -X POST "${AIRESEARCHOS_BASE_URL:-https://airesearchos.com}/api/v1/research" \
-H "Authorization: Bearer $AIRESEARCHOS_API_KEY" \
-H "Content-Type: application/json" \
-d '{"query":"<USER_QUERY>","mode":"<MODE>","reportLength":"standard","skipClarifyingQuestions":false}'
Response when skipClarifyingQuestions=false may include clarifyingQuestions array (status: "clarifying"). Response when skipClarifyingQuestions=true returns status: "queued".
Fields: { id, status, creditsCharged, creditsRemaining, clarifyingQuestions? }
First ensure x402 dependencies are installed:
if [ ! -d "{baseDir}/scripts/node_modules" ]; then
cd {baseDir}/scripts && npm install
fi
Then submit with the x402 helper. Each mode has a separate endpoint:
| Mode | Endpoint | Max Payment | |------|----------|-------------| | scan | /api/x402/research/scan | 0.50 | | dueDiligence | /api/x402/research/due-diligence | 1.50 | | missionCritical | /api/x402/research/mission-critical | 5.00 |
node {baseDir}/scripts/x402-request.mjs \
--url "${AIRESEARCHOS_BASE_URL:-https://airesearchos.com}/api/x402/research/<ENDPOINT_SLUG>" \
--method POST \
--body '{"query":"<USER_QUERY>","reportLength":"standard"}' \
--max-payment <MAX_USDC>
Response: { id, status, mode, query, paymentMethod: "x402", createdAt }
x402 automatically skips clarifying questions. For more targeted results, include relevant context in the query.
CRITICAL: Do NOT poll inline. Do NOT loop. Do NOT run poll-research.mjs. Do NOT run repeated curl commands. Use the cron tool.
After the POST request returns with the research ID, do TWO things:
Step 1: Tell the user the research is submitted:
> Research submitted! > - ID:
Step 2: Call the cron tool with action add to schedule a background status check in 2 minutes.
Build the exec command for the cron payload. The script reads AIRESEARCHOS_API_KEY from the environment automatically (injected by OpenClaw). No secrets in CLI arguments.
API key path:
node {baseDir}/scripts/check-status.mjs --id "<REQUEST_ID>" --base-url "${AIRESEARCHOS_BASE_URL:-https://airesearchos.com}" --api-path "/api/v1"
x402 path:
node {baseDir}/scripts/check-status.mjs --id "<REQUEST_ID>" --base-url "${AIRESEARCHOS_BASE_URL:-https://airesearchos.com}" --api-path "/api/x402"
Then call the cron tool:
Tool: cron
Action: add
Parameters:
name: "AIresearchOS check <REQUEST_ID>"
schedule:
kind: "at"
at: "<ISO_8601_TIMESTAMP_2_MINUTES_FROM_NOW>"
sessionTarget: "isolated"
wakeMode: "now"
payload:
kind: "agentTurn"
message: "Run: <EXEC_COMMAND_FROM_ABOVE> — The script outputs JSON. If action is 'completed', announce the report to the user. If action is 'failed', announce the error. If action is 'pending', schedule another cron check in 2 minutes."
delivery:
mode: "announce"
bestEffort: true
deleteAfterRun: true
Then STOP your turn. The cron job will run in the background and announce results to the user when ready.
If the user asks to check status manually (e.g., "check my research", "is it done?"), run the check-status script:
node {baseDir}/scripts/check-status.mjs --id "<ID>" --base-url "${AIRESEARCHOS_BASE_URL:-https://airesearchos.com}" --api-path "<API_PATH>"
Use --api-path "/api/v1" for API key or --api-path "/api/x402" for x402. The script reads the API key from the environment automatically.
The script outputs JSON with an action field:
action: "completed" — includes the full report. Present it to the user.action: "pending" — includes status, progress, currentStep. Tell the user the progress and suggest checking back.action: "failed" — includes error. Tell the user what went wrong.If status is "clarifying" after submission, present ALL questions to the user at once:
> The research system has follow-up questions: > 1. [question 1] > 2. [question 2] > 3. [question 3] > > Please answer all questions. (Or say "skip" to start without answers.)
Collect answers and submit:
curl -s -X POST "${AIRESEARCHOS_BASE_URL:-https://airesearchos.com}/api/v1/research/<ID>/clarify" \
-H "Authorization: Bearer $AIRESEARCHOS_API_KEY" \
-H "Content-Type: application/json" \
-d '{"answers":["<ANSWER_1>","<ANSWER_2>","<ANSWER_3>"]}'
Response: { id, status: "queued", message }. Then schedule a background cron check (see "Background Status Monitoring" above).
If user says "skip": re-submit with skipClarifyingQuestions=true or pass empty answers.
curl -s "${AIRESEARCHOS_BASE_URL:-https://airesearchos.com}/api/v1/credits" \
-H "Authorization: Bearer $AIRESEARCHOS_API_KEY"
Response: { daily: { allocated, used, remaining, resetsAt }, purchased: { balance }, totalAvailable }
curl -s "${AIRESEARCHOS_BASE_URL:-https://airesearchos.com}/api/v1/research?limit=20&sort=created_at:desc" \
-H "Authorization: Bearer $AIRESEARCHOS_API_KEY"
Response: { data: [...], pagination: { total, limit, offset, hasMore } }
If you already know a research ID is completed:
...
安装 AIresearchOS 后,可以对 AI 说这些话来触发它
Help me get started with AIresearchOS
Explains what AIresearchOS does, walks through the setup, and runs a quick demo based on your current project
Use AIresearchOS to deep research powered by AIresearchOS
Invokes AIresearchOS with the right parameters and returns the result directly in the conversation
What can I do with AIresearchOS in my data & analytics workflow?
Lists the top use cases for AIresearchOS, with example commands for each scenario
将技能文件夹放到 ~/.claude/skills/openclaw-airesearchos/ 目录(个人级,所有项目可用),或 .claude/skills/openclaw-airesearchos/(项目级)。重启 AI 客户端后,用 /openclaw-airesearchos 主动调用,或让 AI 根据上下文自动发现并使用。
AIresearchOS 支持 Claude、Cursor、OpenClaw,可与这些 AI 平台无缝集成,扩展其能力。
AIresearchOS 可免费安装使用。请查阅仓库了解许可证信息。
Deep research powered by AIresearchOS. Submit, track, and retrieve research with clarifying questions. Supports API key auth and x402 USDC payments.
AIresearchOS 属于「Data & Analytics」分类,该分类的技能帮助 AI 智能体在此领域执行专业任务。
Automate my data & analytics tasks using AIresearchOS
Identifies repetitive steps in your workflow and sets up AIresearchOS to handle them automatically