cue-data-mcp
Use when the user wants an external AI agent to query Cue's public MCP data services — regulatory, macro, disclosures, market data, legal full text, holdings…
Skill by sensedeal
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/sensedeal-cue-data-mcp
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: data, research, legal, agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Python package
Network hosts it talks to 2
codeload.github.com· found in install commandcuecue.cn· found in code
Files it touches
~/.claude/skills/cue-data-mcp
What's inside
- Skill name
cue-data-mcp- Description
- Use when the user wants an external AI agent to query Cue's public MCP data services — regulatory, macro, disclosures, market data, legal full text, holdings, entity data, academic, IPO, ESOP, buyback, footnote details — by discovering and connecting to live MCP endpoints.
- License
- MIT
- Metadata
- version: 0.3.6
- Instructions
- 32 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
Files (6)
- README.md
- README.zh-CN.md
- SKILL.md
- SKILL.zh-CN.md
- setup.md · references/
- test_skill_regression.py · scripts/
SKILL.md (first lines)
# Cue Data MCP Use Cue's public MCP data catalog to answer research questions with live external data. The catalog API and the MCP endpoint schemas are authoritative; this skill never implements an MCP client or hardcodes endpoints, domains, or tool names. ## Core flow 1. **Discover live domains.** Call `GET https://cuecue.cn/api/mcp-catalog` — anonymous, no key required. It returns the current domains with `external_status` (`live` / `coming_soon`), `tool_count`, `example_prompts`, and — for live domains — the full `routing` contract. 2. **Match the user's topic to domains.** Pick only the domains that cover the question (e.g. sanctions/KYC → `regulatory`; CPI/GDP → `macro`; SEC filings → `disclosure`; IPO → `ipo`). This catalog covers **15 data domains**; the `omni-reader` (document/web parsing) domain is out of scope here — use the dedicated cue-omni-reader skill for that. 3. **Connect through the routing DTO.** Use the live domain's `routing` verbatim: `url` (e.g. `https://mcp.cuecue.cn/api/<group>/mcp/`), `transport: streamable-http`, `protocol_version: 2025-03-26`. Send `Authorization: Bearer <CUE_API_KEY>` and the `Accept: application/json, text/event-stream` header (streamable-http returns HTTP 406 without it). 4. **Discover tools per domain with `tools/list`.** Never hardcode tool names — the catalog only projects names/titles for logged-in users, and the served tool set changes over time. `tools/list` on the connected endpoint is the live truth. 5. **Call the schema you actually have.** Obey the active tool schema from `tools/list`; never invent arguments. Prefer the domain with the smallest surface that covers the question; do not connect to all domains at once. ## Rules - **The catalog is the only source of truth for reachability.** Always fetch it at the start; a domain can flip `coming_soon` ↔ `live` and routing values can change. Never reuse a stale connection string. - **One domain per session unless the question spans several.** Each MCP connection advertises dozens of tools; connecting everywhere at once is slow and unnecessary. - **Never paste the API key into chat, command arguments, skill files, logs, or generated JSON.** The user configures `CUE_API_KEY` in their own secret facility; the agent never obtains, stores, or transmits it. If a key has appeared in chat, ask the user to rotate it before continuing. - **Report failures truthfully.** A connection error is not a data-absence claim; a tool error is not a domain-wide outage. Retry only what the error says is retryable. Never report a domain as unsupported because one call failed. ## Granted credits and billing Cue credits are the shared wallet behind the data MCP services (and Omni Reader). New users can try the data services without paying: - every account receives **10 granted credits daily** — at 0.625 credits per data call (current standard), roughly **16 data queries per day**; - new accounts receive a **one-time 500-credit gift** when obtaining `CUE_API_KEY` (~816 queries on day one, including the daily grant); - inviting a new user who registers gives both parties 500 credits. The exact allowances and per-call costs follow the server-side billing policy; if a live value differs from the numbers above, report the live value. When the user has no `CUE_API_KEY` yet, direct them to <https://cuecue.cn/hub/api-key> to register and get the key. Never obtain, store, or transmit the key yourself.
From sensedeal/cue-skills, MIT.
More from sensedeal/cue-skills
- 0.627score
cue-omni-reader
sensedeal
Use when the user wants an external AI agent to parse or understand an HTTP(S) URL or an authorized local document, audio, or video source through Cue Omni…
SkillAFreshness: A★ 6Claude Code +9 - 0.627score
cue-research
sensedeal
Use when the user asks a research question they want Cue to run — against a saved 搭子(buddy) template or as free-form deep research.
SkillAFreshness: A★ 6Claude Code +9 - 0.619score
cue-capital-operations
sensedeal
用 Cue 跑「资本运作」场景的深度研究:多源公开数据交叉、结论带来源。 Run Cue deep research for the "Capital Operations" scenario.
SkillAFreshness: A★ 6Claude Code +9 - 0.619score
cue-research-cli
sensedeal
Cue 深度研究 CLI(@cueai/cue-cli,命令 cue)使用说明。用于行业/公司/人物/政策/合规类深度调研:一句话发起多智能体研究,输出带来源的 Markdown 报告,可落盘、可续接会话追问、可指定搭子模板。涉及「深度调研」「尽调」「行研」「公司研究」「竞品对比」「政策跟踪」「人物核查」等需要多源检…
SkillAFreshness: A★ 6Claude Code +9 - 0.619score
cue-credit-diligence
sensedeal
用 Cue 跑「信贷尽调」场景的深度研究:多源公开数据交叉、结论带来源。 Run Cue deep research for the "Credit Diligence" scenario.
SkillAFreshness: A★ 6Claude Code +9 - 0.619score
cue-cross-border-legal
sensedeal
用 Cue 跑「涉外法律」场景的深度研究:多源公开数据交叉、结论带来源。 Run Cue deep research for the "Cross-border Legal" scenario.
SkillAFreshness: A★ 6Claude Code +9 - 0.619score
cue-deep-verification
sensedeal
用 Cue 跑「深度核查」场景的深度研究:多源公开数据交叉、结论带来源。 Run Cue deep research for the "Deep Verification" scenario.
SkillAFreshness: A★ 6Claude Code +9 - 0.619score
cue-equity-research
sensedeal
用 Cue 跑「投资研究」场景的深度研究:多源公开数据交叉、结论带来源。 Run Cue deep research for the "Equity Research" scenario.
SkillAFreshness: A★ 6Claude Code +9 - 0.619score
cue-financial-deep-read
sensedeal
用 Cue 跑「财报深读」场景的深度研究:多源公开数据交叉、结论带来源。 Run Cue deep research for the "Financial Deep Read" scenario.
SkillAFreshness: A★ 6Claude Code +9 - 0.619score
cue-fund-advisory
sensedeal
用 Cue 跑「基金投顾」场景的深度研究:多源公开数据交叉、结论带来源。 Run Cue deep research for the "Fund Advisory" scenario.
SkillAFreshness: A★ 6Claude Code +9 - 0.619score
cue-global-macro
sensedeal
用 Cue 跑「全球宏观」场景的深度研究:多源公开数据交叉、结论带来源。 Run Cue deep research for the "Global Macro" scenario.
SkillAFreshness: A★ 6Claude Code +9 - 0.619score
cue-industry-research
sensedeal
用 Cue 跑「行业研究」场景的深度研究:多源公开数据交叉、结论带来源。 Run Cue deep research for the "Industry Research" scenario.
SkillAFreshness: A★ 6Claude Code +9
Signals
- Stars
- 6
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Sep 11, 2026 (1 month ago)
- Last release
- Oct 9, 2026 (19 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- Version
- 0.3.6
- License
- MIT
- Repository
- github.com/sensedeal/cue-skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from sensedeal
- 0.627score
cue-omni-reader
sensedeal
Use when the user wants an external AI agent to parse or understand an HTTP(S) URL or an authorized local document, audio, or video source through Cue Omni…
SkillAFreshness: A★ 6Claude Code +9 - 0.627score
cue-research
sensedeal
Use when the user asks a research question they want Cue to run — against a saved 搭子(buddy) template or as free-form deep research.
SkillAFreshness: A★ 6Claude Code +9 - 0.619score
cue-cross-border-legal
sensedeal
用 Cue 跑「涉外法律」场景的深度研究:多源公开数据交叉、结论带来源。 Run Cue deep research for the "Cross-border Legal" scenario.
SkillAFreshness: A★ 6Claude Code +9 - 0.619score
cue-credit-diligence
sensedeal
用 Cue 跑「信贷尽调」场景的深度研究:多源公开数据交叉、结论带来源。 Run Cue deep research for the "Credit Diligence" scenario.
SkillAFreshness: A★ 6Claude Code +9 - 0.619score
cue-research-cli
sensedeal
Cue 深度研究 CLI(@cueai/cue-cli,命令 cue)使用说明。用于行业/公司/人物/政策/合规类深度调研:一句话发起多智能体研究,输出带来源的 Markdown 报告,可落盘、可续接会话追问、可指定搭子模板。涉及「深度调研」「尽调」「行研」「公司研究」「竞品对比」「政策跟踪」「人物核查」等需要多源检…
SkillAFreshness: A★ 6Claude Code +9 - 0.619score
cue-deep-verification
sensedeal
用 Cue 跑「深度核查」场景的深度研究:多源公开数据交叉、结论带来源。 Run Cue deep research for the "Deep Verification" scenario.
SkillAFreshness: A★ 6Claude Code +9
Cite this page
cue-data-mcp skill. Agent Mart. https://agentmart.dev/skills/sensedeal-cue-data-mcp (last updated 2026-10-24)