[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79P9ZF5MSSBY8VJDX3B3WV6":3,"skill-same:01M79P9ZF5MSSBY8VJDX3B3WV6":98},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":46,"commitSha":47,"skillDirectory":48,"forksCount":14,"repositoryHtmlUrl":49,"repositoryHomepageUrl":6,"ownerHtmlUrl":50,"ownerBlogUrl":51,"ownerDescription":52,"updatedAt":53,"install":54,"resources":67},"01M79P9ZF5MSSBY8VJDX3B3WV6","aiq-research",null,"Use when asked to run deep research or AI-Q research through a reachable NVIDIA AI-Q Blueprint backend.","MeghVyas3132\u002FClaude-Code-Skills","MeghVyas3132","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F213562095?v=4","01M79MBRQ3EM1GSWXK3QVWPYK6","Megh Vyas","01M79MBRQRZNC3QS0FS1KC6Z09",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FMeghVyas3132\u002FClaude-Code-Skills\u002Fblob\u002Fmain\u002F_other-tools\u002Fcodex\u002Fnvidia\u002Faiq-research\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: aiq-research\ndescription: |\n  Use when asked to run deep research or AI-Q research through a reachable NVIDIA AI-Q Blueprint backend.\nlicense: Apache-2.0\npermissions:\n  env:\n    - AIQ_SERVER_URL\n  network:\n    - http:\u002F\u002Flocalhost:8000\ncompatibility: |\n  Designed for Claude Code, OpenCode, Codex, and Agent Skills-compatible tools. Requires Python 3.11+ and network\n  access to a running local AI-Q Blueprint server at `http:\u002F\u002Flocalhost:8000` by default. Non-local backends must be\n  explicitly trusted by the user and granted by the host tool outside this public skill.\nmetadata:\n  version: \"2.1.0\"\n  author: \"NVIDIA AI-Q Blueprint Team \u003Caiq-blueprint@nvidia.com>\"\n  github-url: \"https:\u002F\u002Fgithub.com\u002FNVIDIA-AI-Blueprints\u002Faiq\"\n  tags:\n    - nvidia\n    - aiq\n    - blueprint\n    - deep-research\n    - research-agents\n    - agent-skills\n  languages:\n    - python\n    - bash\n  domain: \"research-agents\"\nallowed-tools: Read Bash\n---\n\n# AIQ Research Skill\n\n## Purpose\n\nUse this skill to call a locally running NVIDIA AI-Q Blueprint server through the helper script at\n`scripts\u002Faiq.py`.\n\nUse this skill for research-shaped requests, including:\n\n- \"deep research on ...\"\n- \"AIQ research ...\"\n- \"research ...\"\n- \"use AI-Q to answer ...\"\n- \"ask AI-Q about ...\"\n\nDo not use this skill for install, deploy, start, stop, UI, CLI, Docker, Helm, or troubleshooting requests. Those\nbelong to `aiq-deploy`.\n\n## Prerequisites\n\nUsers need:\n\n- Python 3.11+ available as `python3`.\n- A reachable local or self-hosted AI-Q Blueprint backend.\n- `AIQ_SERVER_URL` set when the backend is not running at `http:\u002F\u002Flocalhost:8000`; non-local values must be trusted by\n  the user before any query is sent.\n- A backend configured with authentication disabled for this public helper, or a separate authenticated AI-Q skill for\n  authenticated environments.\n- Network access from the local machine to the AI-Q backend URL.\n- Credentials configured in the backend environment, not in this skill. This public helper does not collect or manage\n  API keys.\n\nThe helper script has no third-party Python package dependencies; it uses Python standard-library HTTP modules.\n\n## Instructions\n\n1. Resolve the target backend URL.\n2. Run `health` before sending research requests.\n3. If no backend is reachable, ask for a backend URL or hand off to `aiq-deploy`.\n4. Before sending any user query, state the exact AI-Q backend URL that will receive it. For non-local URLs, continue\n   only if the user has explicitly confirmed that URL is trusted in the current conversation.\n5. Poll asynchronous deep research jobs when AI-Q returns a job ID.\n6. Present returned reports with citations and source URLs intact.\n7. Stop on failed jobs and show the returned error; do not retry automatically.\n\n### Step 1 - Resolve the backend\n\nUse `AIQ_SERVER_URL` when set. Otherwise try the default local backend:\n\n```bash\npython3 $SKILL_DIR\u002Fscripts\u002Faiq.py health\n```\n\nExpected output: JSON from a reachable AI-Q health endpoint.\n\nIf `health` fails and no explicit `AIQ_SERVER_URL` was set, ask:\n\n```text\nI do not see a reachable local AI-Q backend. Do you already have an AI-Q backend URL you want to use, or should I deploy a local Skill backend?\n```\n\n- If the user provides a URL, set `AIQ_SERVER_URL` for subsequent helper calls and rerun `health`.\n- If the user wants local deployment, hand off to `aiq-deploy` and preserve the original research request.\n- If a reachable backend returns `401` or `403`, stop and explain that this public skill does not manage\n  authentication. Ask the user to use an authenticated AI-Q skill or configure authentication for their environment.\n- If `health` succeeds but `\u002Fchat` or `\u002Fv1\u002Fjobs\u002Fasync\u002Fagents` fails, report that the backend is reachable but not\n  compatible with this public research flow, then offer to run `aiq-deploy` validation.\n\n### Step 2 - Send the routed research request\n\nBefore sending the request, state the resolved endpoint:\n\n```text\nI will send this query to \u003CAIQ_SERVER_URL>. Make sure this endpoint is trusted before sending sensitive information.\n```\n\nDo not send credentials, cookies, bearer tokens, or secret values through the query text.\n\nRun:\n\n```bash\npython3 $SKILL_DIR\u002Fscripts\u002Faiq.py chat \"\u003CUSER_QUESTION>\"\n```\n\nExpected output:\n\n- A normal JSON response for shallow or direct answers.\n- Or structured JSON containing `{\"status\": \"deep_research_running\", \"job_id\": \"\u003CJOB_ID>\"}` for asynchronous deep\n  research.\n\nIf the response is normal JSON, present the result immediately. Do not force polling when there is no `job_id`.\n\n### Step 3 - Poll asynchronous jobs\n\nIf the response includes `deep_research_running`, extract the `job_id` and poll with the same absolute script path:\n\n```bash\npython3 $SKILL_DIR\u002Fscripts\u002Faiq.py research_poll \u003CJOB_ID>\n```\n\nExpected output: the final report JSON when the job completes successfully.\n\nUse the runtime's non-blocking or background execution mechanism when available. If the chosen execution method requires\nescalated permissions, request explicit user approval first and explain why. Tell the user that deep research is running\nin the background.\n\n### Step 4 - Resume after interruptions\n\nIf polling is interrupted, the job continues server-side. Resume with:\n\n```bash\npython3 $SKILL_DIR\u002Fscripts\u002Faiq.py status \u003CJOB_ID>\npython3 $SKILL_DIR\u002Fscripts\u002Faiq.py report \u003CJOB_ID>\npython3 $SKILL_DIR\u002Fscripts\u002Faiq.py research_poll \u003CJOB_ID>\n```\n\nUse `status` to inspect job status and saved artifacts. Use `report` when the job has already finished and you only need\nthe final output. Use `research_poll` to keep waiting for completion.\n\n### Step 5 - Present the report\n\nWhen `research_poll` completes successfully, fetch and present the full report. Keep citations and source URLs intact.\nIf the job status is `failed`, `failure`, or `cancelled`, show the error from the status response and ask whether the\nuser wants to retry with a narrower query or different approach.\n\n## Version Compatibility\n\n**IMPORTANT:** This skill is designed for NVIDIA AI-Q Blueprint version 2.1.0.\n\nSemantic Versioning Compatibility Rules:\n\n```text\nSkill version: X.Y.Z\nBlueprint or endpoint version: A.B.C\n\nCompatible IF:\n1. A == X (Major versions MUST match)\n2. B >= Y (Minor version must be equal or greater)\n3. C can be anything (Patch version does not affect compatibility)\n```\n\nExamples:\n\n- Skill version 2.1.0 is compatible with Blueprint version 2.1.0.\n- Skill version 2.1.0 is compatible with Blueprint version 2.2.0.\n- Skill version 2.1.0 is compatible with Blueprint version 2.1.5.\n- Skill version 2.1.0 is not compatible with Blueprint version 3.0.0.\n- Skill version 2.1.0 is not compatible with Blueprint version 2.0.0.\n\nIf your Blueprint version is not compatible:\n\n1. Check for an updated skill version matching your Blueprint version.\n2. Use a Blueprint version compatible with this skill.\n3. Proceed with caution only when the user accepts the compatibility risk; API routes or response shapes may have\n   changed.\n\n## Available Scripts\n\n| Script | Purpose | Arguments |\n|---|---|---|\n| `scripts\u002Faiq.py health` | Check whether the configured server responds | none |\n| `scripts\u002Faiq.py chat` | POST `\u002Fchat`; may return inline output or a deep-research job ID | `\u003Cquery>` |\n| `scripts\u002Faiq.py agents` | List available async agent types | none |\n| `scripts\u002Faiq.py submit` | Submit an explicit async job | `\u003Cquery> [agent_type]` |\n| `scripts\u002Faiq.py research` | Submit an async job, poll, and print the final report JSON | `\u003Cquery> [agent_type]` |\n| `scripts\u002Faiq.py research_poll` | Resume polling an existing async job | `\u003Cjob_id>` |\n| `scripts\u002Faiq.py status` | Fetch job status plus `\u002Fstate` artifacts | `\u003Cjob_id>` |\n| `scripts\u002Faiq.py state` | Fetch event-store artifacts only | `\u003Cjob_id>` |\n| `scripts\u002Faiq.py report` | Fetch the final report for a completed job | `\u003Cjob_id>` |\n| `scripts\u002Faiq.py stream` | Stream SSE events from a job | `\u003Cjob_id>` |\n| `scripts\u002Faiq.py cancel` | Cancel a running job | `\u003Cjob_id>` |\n\nWhen the host supports a `run_script()` helper, call it with `scripts\u002Faiq.py` and the arguments above. Otherwise, run\nthe equivalent shell command, such as `python3 $SKILL_DIR\u002Fscripts\u002Faiq.py health`.\n\n## Environment Variables\n\n| Variable | Required | Default | Description |\n|---|---:|---|---|\n| `AIQ_SERVER_URL` | No | `http:\u002F\u002Flocalhost:8000` | Local or self-hosted AI-Q server base URL |\n\n## Security Best Practices\n\n- Do not put API keys, bearer tokens, cookies, or basic-auth credentials in `AIQ_SERVER_URL`.\n- Store backend credentials in the AI-Q deployment environment, not in this skill or command examples.\n- User query text is transmitted to the configured `AIQ_SERVER_URL`. Confirm the endpoint is trusted before sending\n  sensitive or confidential information.\n- Treat returned reports as potentially sensitive if the backend uses private data sources.\n- Do not truncate citations or source URLs from returned reports.\n\n## Limitations\n\n- This skill requires a running AI-Q backend; it does not deploy one.\n- The public helper does not manage authentication tokens or cookies.\n- Remote `AIQ_SERVER_URL` endpoints may log prompts, responses, and metadata.\n- If the backend returns HTTP 500 or lacks async agents, report the failure instead of fabricating a research answer.\n\n## Examples\n\n### Example 1: Run a routed chat or research request\n\n```bash\npython3 $SKILL_DIR\u002Fscripts\u002Faiq.py health\npython3 $SKILL_DIR\u002Fscripts\u002Faiq.py chat \"Compare local AIQ deep research with a standard web search workflow\"\n```\n\nExpected output:\n\n```text\n\u003Chealth JSON from AI-Q>\n\u003CJSON chat response or {\"status\": \"deep_research_running\", \"job_id\": \"\u003CJOB_ID>\"}>\n```\n\nIf AI-Q returns a job ID, continue with `research_poll`.\n\n### Example 2: Resume an existing job\n\n```bash\npython3 $SKILL_DIR\u002Fscripts\u002Faiq.py status \u003CJOB_ID>\npython3 $SKILL_DIR\u002Fscripts\u002Faiq.py research_poll \u003CJOB_ID>\n```\n\nReplace `\u003CJOB_ID>` with the UUID returned by AI-Q. Expected output: status JSON followed by the report JSON when the\njob completes. If the job failed, show the returned status and do not retry automatically.\n\n## References\n\n| Topic | Documentation |\n|---|---|\n| Helper script | `scripts\u002Faiq.py` |\n| Deployment and backend validation | `..\u002Faiq-deploy\u002FSKILL.md` |\n\n## Common Issues\n\n### Issue: No backend is reachable\n\n**Symptoms:**\n\n- `health` fails with connection refused.\n- The default `http:\u002F\u002Flocalhost:8000` URL does not respond.\n\n**Causes:**\n\n- AI-Q is not running.\n- AI-Q is running on a different host or port.\n- A local firewall or network setting blocks the connection.\n\n**Solutions:**\n\n1. Ask whether the user has an existing AI-Q backend URL.\n2. If they provide one, set it and rerun health:\n   ```bash\n   export AIQ_SERVER_URL=\"http:\u002F\u002Flocalhost:\u003CPORT>\"\n   python3 $SKILL_DIR\u002Fscripts\u002Faiq.py health\n   ```\n3. If they want a local backend, hand off to `aiq-deploy` and preserve the original research request.\n\n### Issue: Backend requires authentication\n\n**Symptoms:**\n\n- Requests fail with HTTP 401 or HTTP 403.\n- The backend is reachable but rejects `\u002Fchat` or async job calls.\n\n**Causes:**\n\n- The backend was deployed with authentication enabled.\n- The public helper does not attach user tokens or cookies.\n\n**Solutions:**\n\n1. Stop and explain that this public skill does not manage authentication.\n2. Ask the user to use an authenticated AI-Q skill or configure their backend for this public local workflow.\n3. Rerun `health` and the original query only after the authentication boundary is resolved.\n\n### Issue: Health succeeds but research routes fail\n\n**Symptoms:**\n\n- `health` returns successfully.\n- `\u002Fchat`, `\u002Fv1\u002Fjobs\u002Fasync\u002Fagents`, or polling commands fail.\n\n**Causes:**\n\n- The backend is not using an API-enabled AI-Q config.\n- The async job registry is not available in the selected backend.\n- The backend version is incompatible with this skill.\n\n**Solutions:**\n\n1. Run:\n   ```bash\n   python3 $SKILL_DIR\u002Fscripts\u002Faiq.py agents\n   ```\n2. If agents are unavailable, report the compatibility failure and offer to run `aiq-deploy` validation.\n3. Confirm the deployed Blueprint version is compatible with skill version 2.1.0.\n\n### Issue: Job is interrupted or appears stuck\n\n**Symptoms:**\n\n- Local polling is interrupted.\n- The job keeps showing `running`.\n- Poll output shows `running`, but a report is returned or cancel says the job is already `success`.\n\n**Causes:**\n\n- Deep research is asynchronous and continues server-side.\n- Local polling output can lag behind terminal server state.\n\n**Solutions:**\n\n1. Check current state:\n   ```bash\n   python3 $SKILL_DIR\u002Fscripts\u002Faiq.py status \u003CJOB_ID>\n   ```\n2. If `has_report: true` or `job_status.status: success`, fetch the report:\n   ```bash\n   python3 $SKILL_DIR\u002Fscripts\u002Faiq.py report \u003CJOB_ID>\n   ```\n3. If the job is still running, continue polling:\n   ```bash\n   python3 $SKILL_DIR\u002Fscripts\u002Faiq.py research_poll \u003CJOB_ID>\n   ```\n","# AIQ Research Skill\n\n## Purpose\n\nUse this skill to call a locally running NVIDIA AI-Q Blueprint server through the helper script at\n`scripts\u002Faiq.py`.\n\nUse this skill for research-shaped requests, including:\n\n- \"deep research on ...\"\n- \"AIQ research ...\"\n- \"research ...\"\n- \"use AI-Q to answer ...\"\n- \"ask AI-Q about ...\"\n\nDo not use this skill for install, deploy, start, stop, UI, CLI, Docker, Helm, or troubleshooting requests. Those\nbelong to `aiq-deploy`.\n\n## Prerequisites\n\nUsers need:\n\n- Python 3.11+ available as `python3`.\n- A reachable local or self-hosted AI-Q Blueprint backend.\n- `AIQ_SERVER_URL` set when the backend is not running at `http:\u002F\u002Flocalhost:8000`; non-local values must be trusted by\n  the user before any query is sent.\n- A backend configured with authentication disabled for this public helper, or a separate authenticated AI-Q skill for\n  authenticated environments.\n- Network access from the local machine to the AI-Q backend URL.\n- Credentials configured in the backend environment, not in this skill. This public helper does not collect or manage\n  API keys.\n\nThe helper script has no third-party Python package dependencies; it uses Python standard-library HTTP modules.\n\n## Instructions\n\n1. Resolve the target backend URL.\n2. Run `health` before sending research requests.\n3. If no backend is reachable, ask for a backend URL or hand off to `aiq-deploy`.\n4. Before sending any user query, state the exact AI-Q backend URL that will receive it. For non-local URLs, continue\n   only if the user has explicitly confirmed that URL is trusted in the current conversation.\n5. Poll asynchronous deep research jobs when AI-Q returns a job ID.\n6. Present returned reports with citations and source URLs intact.\n7. Stop on failed jobs and show the returned error; do not retry automatically.\n\n### Step 1 - Resolve the backend\n\nUse `AIQ_SERVER_URL` when set. Otherwise try the default local backend:\n\n```bash\npython3 $SKILL_DIR\u002Fscripts\u002Faiq.py health\n```\n\nExpected output: JSON from a reachable AI-Q health endpoint.\n\nIf `health` fails and no explicit `AIQ_SERVER_URL` was set, ask:\n\n```text\nI do not see a reachable local AI-Q backend. Do you already have an AI-Q backend URL you want to use, or should I deploy a local Skill backend?\n```\n\n- If the user provides a URL, set `AIQ_SERVER_URL` for subsequent helper calls and rerun `health`.\n- If the user wants local deployment, hand off to `aiq-deploy` and preserve the original research request.\n- If a reachable backend returns `401` or `403`, stop and explain that this public skill does not manage\n  authentication. Ask the user to use an authenticated AI-Q skill or configure authentication for their environment.\n- If `health` succeeds but `\u002Fchat` or `\u002Fv1\u002Fjobs\u002Fasync\u002Fagents` fails, report that the backend is reachable but not\n  compatible with this public research flow, then offer to run `aiq-deploy` validation.\n\n### Step 2 - Send the routed research request\n\nBefore sending the request, state the resolved endpoint:\n\n```text\nI will send this query to \u003CAIQ_SERVER_URL>. Make sure this endpoint is trusted before sending sensitive information.\n```\n\nDo not send credentials, cookies, bearer tokens, or secret values through the query text.\n\nRun:\n\n```bash\npython3 $SKILL_DIR\u002Fscripts\u002Faiq.py chat \"\u003CUSER_QUESTION>\"\n```\n\nExpected output:\n\n- A normal JSON response for shallow or direct answers.\n- Or structured JSON containing `{\"status\": \"deep_research_running\", \"job_id\": \"\u003CJOB_ID>\"}` for asynchronous deep\n  research.\n\nIf the response is normal JSON, present the result immediately. Do not force polling when there is no `job_id`.\n\n### Step 3 - Poll asynchronous jobs\n\nIf the response includes `deep_research_running`, extract the `job_id` and poll with the same absolute script path:\n\n```bash\npython3 $SKILL_DIR\u002Fscripts\u002Faiq.py research_poll \u003CJOB_ID>\n```\n\nExpected output: the final report JSON when the job completes successfully.\n\nUse the runtime's non-blocking or background execution mechanism when available. If the chosen execution method requires\nescalated permissions, request explicit user approval first and explain why. Tell the user that deep research is running\nin the background.\n\n### Step 4 - Resume after interruptions\n\nIf polling is interrupted, the job continues server-side. Resume with:\n\n```bash\npython3 $SKILL_DIR\u002Fscripts\u002Faiq.py status \u003CJOB_ID>\npython3 $SKILL_DIR\u002Fscripts\u002Faiq.py report \u003CJOB_ID>\npython3 $SKILL_DIR\u002Fscripts\u002Faiq.py research_poll \u003CJOB_ID>\n```\n\nUse `status` to inspect job status and saved artifacts. Use `report` when the job has already finished and you only need\nthe final output. Use `research_poll` to keep waiting for completion.\n\n### Step 5 - Present the report\n\nWhen `research_poll` completes successfully, fetch and present the full report. Keep citations and source URLs intact.\nIf the job status is `failed`, `failure`, or `cancelled`, show the error from the status response and ask whether the\nuser wants to retry with a narrower query or different approach.\n\n## Version Compatibility\n\n**IMPORTANT:** This skill is designed for NVIDIA AI-Q Blueprint version 2.1.0.\n\nSemantic Versioning Compatibility Rules:\n\n```text\nSkill version: X.Y.Z\nBlueprint or endpoint version: A.B.C\n\nCompatible IF:\n1. A == X (Major versions MUST match)\n2. B >= Y (Minor version must be equal or greater)\n3. C can be anything (Patch version does not affect compatibility)\n```\n\nExamples:\n\n- Skill version 2.1.0 is compatible with Blueprint version 2.1.0.\n- Skill version 2.1.0 is compatible with Blueprint version 2.2.0.\n- Skill version 2.1.0 is compatible with Blueprint version 2.1.5.\n- Skill version 2.1.0 is not compatible with Blueprint version 3.0.0.\n- Skill version 2.1.0 is not compatible with Blueprint version 2.0.0.\n\nIf your Blueprint version is not compatible:\n\n1. Check for an updated skill version matching your Blueprint version.\n2. Use a Blueprint version compatible with this skill.\n3. Proceed with caution only when the user accepts the compatibility risk; API routes or response shapes may have\n   changed.\n\n## Available Scripts\n\n| Script | Purpose | Arguments |\n|---|---|---|\n| `scripts\u002Faiq.py health` | Check whether the configured server responds | none |\n| `scripts\u002Faiq.py chat` | POST `\u002Fchat`; may return inline output or a deep-research job ID | `\u003Cquery>` |\n| `scripts\u002Faiq.py agents` | List available async agent types | none |\n| `scripts\u002Faiq.py submit` | Submit an explicit async job | `\u003Cquery> [agent_type]` |\n| `scripts\u002Faiq.py research` | Submit an async job, poll, and print the final report JSON | `\u003Cquery> [agent_type]` |\n| `scripts\u002Faiq.py research_poll` | Resume polling an existing async job | `\u003Cjob_id>` |\n| `scripts\u002Faiq.py status` | Fetch job status plus `\u002Fstate` artifacts | `\u003Cjob_id>` |\n| `scripts\u002Faiq.py state` | Fetch event-store artifacts only | `\u003Cjob_id>` |\n| `scripts\u002Faiq.py report` | Fetch the final report for a completed job | `\u003Cjob_id>` |\n| `scripts\u002Faiq.py stream` | Stream SSE events from a job | `\u003Cjob_id>` |\n| `scripts\u002Faiq.py cancel` | Cancel a running job | `\u003Cjob_id>` |\n\nWhen the host supports a `run_script()` helper, call it with `scripts\u002Faiq.py` and the arguments above. Otherwise, run\nthe equivalent shell command, such as `python3 $SKILL_DIR\u002Fscripts\u002Faiq.py health`.\n\n## Environment Variables\n\n| Variable | Required | Default | Description |\n|---|---:|---|---|\n| `AIQ_SERVER_URL` | No | `http:\u002F\u002Flocalhost:8000` | Local or self-hosted AI-Q server base URL |\n\n## Security Best Practices\n\n- Do not put API keys, bearer tokens, cookies, or basic-auth credentials in `AIQ_SERVER_URL`.\n- Store backend credentials in the AI-Q deployment environment, not in this skill or command examples.\n- User query text is transmitted to the configured `AIQ_SERVER_URL`. Confirm the endpoint is trusted before sending\n  sensitive or confidential information.\n- Treat returned reports as potentially sensitive if the backend uses private data sources.\n- Do not truncate citations or source URLs from returned reports.\n\n## Limitations\n\n- This skill requires a running AI-Q backend; it does not deploy one.\n- The public helper does not manage authentication tokens or cookies.\n- Remote `AIQ_SERVER_URL` endpoints may log prompts, responses, and metadata.\n- If the backend returns HTTP 500 or lacks async agents, report the failure instead of fabricating a research answer.\n\n## Examples\n\n### Example 1: Run a routed chat or research request\n\n```bash\npython3 $SKILL_DIR\u002Fscripts\u002Faiq.py health\npython3 $SKILL_DIR\u002Fscripts\u002Faiq.py chat \"Compare local AIQ deep research with a standard web search workflow\"\n```\n\nExpected output:\n\n```text\n\u003Chealth JSON from AI-Q>\n\u003CJSON chat response or {\"status\": \"deep_research_running\", \"job_id\": \"\u003CJOB_ID>\"}>\n```\n\nIf AI-Q returns a job ID, continue with `research_poll`.\n\n### Example 2: Resume an existing job\n\n```bash\npython3 $SKILL_DIR\u002Fscripts\u002Faiq.py status \u003CJOB_ID>\npython3 $SKILL_DIR\u002Fscripts\u002Faiq.py research_poll \u003CJOB_ID>\n```\n\nReplace `\u003CJOB_ID>` with the UUID returned by AI-Q. Expected output: status JSON followed by the report JSON when the\njob completes. If the job failed, show the returned status and do not retry automatically.\n\n## References\n\n| Topic | Documentation |\n|---|---|\n| Helper script | `scripts\u002Faiq.py` |\n| Deployment and backend validation | `..\u002Faiq-deploy\u002FSKILL.md` |\n\n## Common Issues\n\n### Issue: No backend is reachable\n\n**Symptoms:**\n\n- `health` fails with connection refused.\n- The default `http:\u002F\u002Flocalhost:8000` URL does not respond.\n\n**Causes:**\n\n- AI-Q is not running.\n- AI-Q is running on a different host or port.\n- A local firewall or network setting blocks the connection.\n\n**Solutions:**\n\n1. Ask whether the user has an existing AI-Q backend URL.\n2. If they provide one, set it and rerun health:\n   ```bash\n   export AIQ_SERVER_URL=\"http:\u002F\u002Flocalhost:\u003CPORT>\"\n   python3 $SKILL_DIR\u002Fscripts\u002Faiq.py health\n   ```\n3. If they want a local backend, hand off to `aiq-deploy` and preserve the original research request.\n\n### Issue: Backend requires authentication\n\n**Symptoms:**\n\n- Requests fail with HTTP 401 or HTTP 403.\n- The backend is reachable but rejects `\u002Fchat` or async job calls.\n\n**Causes:**\n\n- The backend was deployed with authentication enabled.\n- The public helper does not attach user tokens or cookies.\n\n**Solutions:**\n\n1. Stop and explain that this public skill does not manage authentication.\n2. Ask the user to use an authenticated AI-Q skill or configure their backend for this public local workflow.\n3. Rerun `health` and the original query only after the authentication boundary is resolved.\n\n### Issue: Health succeeds but research routes fail\n\n**Symptoms:**\n\n- `health` returns successfully.\n- `\u002Fchat`, `\u002Fv1\u002Fjobs\u002Fasync\u002Fagents`, or polling commands fail.\n\n**Causes:**\n\n- The backend is not using an API-enabled AI-Q config.\n- The async job registry is not available in the selected backend.\n- The backend version is incompatible with this skill.\n\n**Solutions:**\n\n1. Run:\n   ```bash\n   python3 $SKILL_DIR\u002Fscripts\u002Faiq.py agents\n   ```\n2. If agents are unavailable, report the compatibility failure and offer to run `aiq-deploy` validation.\n3. Confirm the deployed Blueprint version is compatible with skill version 2.1.0.\n\n### Issue: Job is interrupted or appears stuck\n\n**Symptoms:**\n\n- Local polling is interrupted.\n- The job keeps showing `running`.\n- Poll output shows `running`, but a report is returned or cancel says the job is already `success`.\n\n**Causes:**\n\n- Deep research is asynchronous and continues server-side.\n- Local polling output can lag behind terminal server state.\n\n**Solutions:**\n\n1. Check current state:\n   ```bash\n   python3 $SKILL_DIR\u002Fscripts\u002Faiq.py status \u003CJOB_ID>\n   ```\n2. If `has_report: true` or `job_status.status: success`, fetch the report:\n   ```bash\n   python3 $SKILL_DIR\u002Fscripts\u002Faiq.py report \u003CJOB_ID>\n   ```\n3. If the job is still running, continue polling:\n   ```bash\n   python3 $SKILL_DIR\u002Fscripts\u002Faiq.py research_poll \u003CJOB_ID>\n   ```\n",{"name":5,"license":15,"metadata":24,"description":38,"permissions":39,"allowed-tools":44,"compatibility":45},{"tags":25,"author":32,"domain":30,"version":33,"languages":34,"github-url":37},[26,27,28,29,30,31],"nvidia","aiq","blueprint","deep-research","research-agents","agent-skills","NVIDIA AI-Q Blueprint Team \u003Caiq-blueprint@nvidia.com>","2.1.0",[35,36],"python","bash","https:\u002F\u002Fgithub.com\u002FNVIDIA-AI-Blueprints\u002Faiq","Use when asked to run deep research or AI-Q research through a reachable NVIDIA AI-Q Blueprint backend.\n",{"env":40,"network":42},[41],"AIQ_SERVER_URL",[43],"http:\u002F\u002Flocalhost:8000","Read Bash","Designed for Claude Code, OpenCode, Codex, and Agent Skills-compatible tools. Requires Python 3.11+ and network\naccess to a running local AI-Q Blueprint server at `http:\u002F\u002Flocalhost:8000` by default. Non-local backends must be\nexplicitly trusted by the user and granted by the host tool outside this public skill.\n","main","12a76cdd6f897f0872b282adeec07c7a07cf6bcc","_other-tools\u002Fcodex\u002Fnvidia\u002Faiq-research","https:\u002F\u002Fgithub.com\u002FMeghVyas3132\u002FClaude-Code-Skills","https:\u002F\u002Fgithub.com\u002FMeghVyas3132","https:\u002F\u002Fmegh.tech","Entrepreneur \u002F Platform Engineer \u002F DevOps \u002F Backend developer \u002F System Architect \u002F Salesforce admin-dev \u002F AI Engineer \u002F Junior SRE","2026-07-25 05:29:37",{"source":8,"skillName":5,"commands":55,"prompt":65,"zipAvailable":66},[56,59,62],{"tool":57,"command":58},"npx","npx skills add MeghVyas3132\u002FClaude-Code-Skills --skill aiq-research",{"tool":60,"command":61},"bunx","bunx skills add MeghVyas3132\u002FClaude-Code-Skills --skill aiq-research",{"tool":63,"command":64},"pnpm","pnpm dlx skills add MeghVyas3132\u002FClaude-Code-Skills --skill aiq-research","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FMeghVyas3132\u002FClaude-Code-Skills\u002Fblob\u002Fmain\u002F_other-tools\u002Fcodex\u002Fnvidia\u002Faiq-research\u002FSKILL.md",true,[68,74,78,82,88,92],{"path":69,"resourceType":70,"resourceTypeDesc":71,"sizeBytes":72,"sourceHtmlUrl":73},"agents\u002Fopenai.yaml","REFERENCE","参考文档",169,"https:\u002F\u002Fgithub.com\u002FMeghVyas3132\u002FClaude-Code-Skills\u002Fblob\u002Fmain\u002F_other-tools\u002Fcodex\u002Fnvidia\u002Faiq-research\u002Fagents\u002Fopenai.yaml",{"path":75,"resourceType":70,"resourceTypeDesc":71,"sizeBytes":76,"sourceHtmlUrl":77},"BENCHMARK.md",3807,"https:\u002F\u002Fgithub.com\u002FMeghVyas3132\u002FClaude-Code-Skills\u002Fblob\u002Fmain\u002F_other-tools\u002Fcodex\u002Fnvidia\u002Faiq-research\u002FBENCHMARK.md",{"path":79,"resourceType":70,"resourceTypeDesc":71,"sizeBytes":80,"sourceHtmlUrl":81},"evals\u002Fevals.json",2393,"https:\u002F\u002Fgithub.com\u002FMeghVyas3132\u002FClaude-Code-Skills\u002Fblob\u002Fmain\u002F_other-tools\u002Fcodex\u002Fnvidia\u002Faiq-research\u002Fevals\u002Fevals.json",{"path":83,"resourceType":84,"resourceTypeDesc":85,"sizeBytes":86,"sourceHtmlUrl":87},"scripts\u002Faiq.py","SCRIPT","脚本",18473,"https:\u002F\u002Fgithub.com\u002FMeghVyas3132\u002FClaude-Code-Skills\u002Fblob\u002Fmain\u002F_other-tools\u002Fcodex\u002Fnvidia\u002Faiq-research\u002Fscripts\u002Faiq.py",{"path":89,"resourceType":70,"resourceTypeDesc":71,"sizeBytes":90,"sourceHtmlUrl":91},"skill-card.md",3766,"https:\u002F\u002Fgithub.com\u002FMeghVyas3132\u002FClaude-Code-Skills\u002Fblob\u002Fmain\u002F_other-tools\u002Fcodex\u002Fnvidia\u002Faiq-research\u002Fskill-card.md",{"path":93,"resourceType":94,"resourceTypeDesc":95,"sizeBytes":96,"sourceHtmlUrl":97},"skill.oms.sig","OTHER","其他",4805,"https:\u002F\u002Fgithub.com\u002FMeghVyas3132\u002FClaude-Code-Skills\u002Fblob\u002Fmain\u002F_other-tools\u002Fcodex\u002Fnvidia\u002Faiq-research\u002Fskill.oms.sig",[99,107,115,122,129,136],{"skillNo":100,"name":101,"nameZh":6,"description":102,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":103,"sourceHtmlUrl":104,"categories":105,"keywords":106,"publishedAt":19},"01M79QXVKBA07BMENY2S4C9B0D","skill-creator","Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.","https:\u002F\u002Fgithub.com\u002FMeghVyas3132\u002FClaude-Code-Skills\u002Fblob\u002Fmain\u002Fskill-creator\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002FMeghVyas3132\u002FClaude-Code-Skills\u002Fblob\u002Fmain\u002Fskill-creator\u002FSKILL.md",[],[],{"skillNo":108,"name":109,"nameZh":6,"description":110,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":111,"licenseFileUrl":6,"sourceHtmlUrl":112,"categories":113,"keywords":114,"publishedAt":19},"01M79QXPTA7529Z7QX9TYC9X8M","skill-audit","Pre-install security scanner for AI agent skills. 7.5% of 14,706 skills are malicious. Audit before you trust.","MIT","https:\u002F\u002Fgithub.com\u002FMeghVyas3132\u002FClaude-Code-Skills\u002Fblob\u002Fmain\u002Fskill-audit\u002FSKILL.md",[],[],{"skillNo":116,"name":117,"nameZh":6,"description":118,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":111,"licenseFileUrl":6,"sourceHtmlUrl":119,"categories":120,"keywords":121,"publishedAt":19},"01M79QWZ2BB5GTX3PCHEEDG3A2","vercel-react-best-practices","React and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React\u002FNext.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance improvements.","https:\u002F\u002Fgithub.com\u002FMeghVyas3132\u002FClaude-Code-Skills\u002Fblob\u002Fmain\u002Freact-best-practices\u002FSKILL.md",[],[],{"skillNo":123,"name":124,"nameZh":6,"description":125,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":126,"categories":127,"keywords":128,"publishedAt":19},"01M79QWAADBR52R570H6N25F47","impeccable","Use when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract, or otherwise improve a frontend interface. Covers websites, landing pages, dashboards, product UI, app shells, components, forms, settings, onboarding, and empty states. Handles UX review, visual hierarchy, information architecture, cognitive load, accessibility, performance, responsive behavior, theming, anti-patterns, typography, fonts, spacing, layout, alignment, color, motion, micro-interactions, UX copy, error states, edge cases, i18n, and reusable design systems or tokens. Also use for bland designs that need to become bolder or more delightful, loud designs that should become quieter, live browser iteration on UI elements, or ambitious visual effects that should feel technically extraordinary. Not for backend-only or non-UI tasks.","https:\u002F\u002Fgithub.com\u002FMeghVyas3132\u002FClaude-Code-Skills\u002Fblob\u002Fmain\u002Fimpeccable\u002FSKILL.md",[],[],{"skillNo":130,"name":131,"nameZh":6,"description":132,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":111,"licenseFileUrl":6,"sourceHtmlUrl":133,"categories":134,"keywords":135,"publishedAt":19},"01M79QTGQJGPXPJDB09WBZHRND","vercel-composition-patterns","React composition patterns that scale. Use when refactoring components with boolean prop proliferation, building flexible component libraries, or designing reusable APIs. Triggers on tasks involving compound components, render props, context providers, or component architecture. Includes React 19 API changes.","https:\u002F\u002Fgithub.com\u002FMeghVyas3132\u002FClaude-Code-Skills\u002Fblob\u002Fmain\u002Fcomposition-patterns\u002FSKILL.md",[],[],{"skillNo":137,"name":138,"nameZh":6,"description":139,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":111,"licenseFileUrl":6,"sourceHtmlUrl":140,"categories":141,"keywords":142,"publishedAt":19},"01M79QT5TCZBBKP5ECWE0QD9PT","bumblebee","Run Bumblebee supply-chain inventory and exposure scans on macOS\u002FLinux to detect compromised packages, extensions, and MCP host configs.","https:\u002F\u002Fgithub.com\u002FMeghVyas3132\u002FClaude-Code-Skills\u002Fblob\u002Fmain\u002Fbumblebee\u002FSKILL.md",[],[]]