The MCP tool directory.

[API] Create a new Railway project ⚡️ Best for: ✓ Starting new applications ✓ Setting up development environments ✓ Creating project spaces ⚠️ Not for: × Duplicating existing projects → Next steps: service_create_from_repo, service_create_from_image, database_deploy → Related: project_delete, project_update

[API] Delete a Railway project and all its resources ⚡️ Best for: ✓ Removing unused projects ✓ Cleaning up test projects ⚠️ Not for: × Temporary project deactivation × Service-level cleanup (use service_delete) → Prerequisites: project_list, project_info → Alternatives: service_delete → Related: project_create

List all environments in a project

[API] Get detailed information about a specific Railway project ⚡️ Best for: ✓ Viewing project details and status ✓ Checking environments and services ✓ Project configuration review → Prerequisites: project_list → Next steps: service_list, variable_list → Related: project_update, project_delete

[API] List all projects in your Railway account ⚡️ Best for: ✓ Getting an overview of all projects ✓ Finding project IDs ✓ Project discovery and management → Next steps: project_info, service_list → Related: project_create, project_delete

Get details about specific GitHub Projects resources. Use this tool to get details about individual projects, project fields, and project items by their unique IDs.

Tools for listing GitHub Projects resources. Use this tool to list projects for a user or organization, or list project fields and items for a specific project.

Add, update, or delete project items, or create status updates in a GitHub Project.

Expensive LLM job — does not spend a researched cluster. Proposes a workstream's articles from its research dossier and writes them into the workstream as status=Idea. Each idea cites the claim ids it is built on, states its angle, takes a distinct shape, and targets a reading level. Ideas that cite no claim, cite only one publisher, repeat a shape or angle already taken, or reach for template phrasing are rejected by code and regenerated once. Existing ideas that anyone has acted on — approved, briefed, annotated or already written — are kept; untouched proposals are replaced. Runs automatically at the end of deep research; call it directly to re-propose from research already paid for.

Returns the IP ranges used by the Ahrefs public web crawler, typically for allowlisting or firewall configuration. Use doc tool first to get the real input schema.

Returns the list of individual IP addresses currently used by the Ahrefs public web crawler. Use doc tool first to get the real input schema.

Publishes all ready articles in a workstream at once. Cross-links between articles in the cluster are preserved (not stripped). Articles must be ReadyForReview and already reviewed in the Claims drawer. Unreviewed drafts are skipped.

Get information on a specific pull request in GitHub repository.

Create and/or submit, delete review of a pull request. Available methods: - create: Create a new review of a pull request. If "event" parameter is provided, the review is submitted. If "event" is omitted, a pending review is created. - submit_pending: Submit an existing pending review of a pull request. This requires that a pending review exists for the current user on the specified pull request. The "body" and "event" parameters are used when submitting the review. - delete_pending: Delete an existing pending review of a pull request. This requires that a pending review exists for the current user on the specified pull request.

Get machine-readable payment instructions to purchase an OpenClaw Intel Pro API key ($9 one-time). Returns multiple payment options (PayPal, crypto) with direct action URLs. Pro key unlocks: full market reports (15+ metrics vs 3), 1000 API calls/day (vs 10 free), historical data access, and growth projections. ROI: One actionable insight from full data can save hours of manual research.

Push multiple files to a GitHub repository in a single commit

Put an active employee on leave (suspend without firing)

Run Python in a Pyodide sandbox with optional PEP 723 requirements.

奇门遁甲（Qimen Dunjia）是中国古代三式之一，被称为"帝王之学"。通过天时、地利、人和三维度分析时空能量场。 ## 核心概念 - **九宫**：坎(1)、坤(2)、震(3)、巽(4)、中(5)、乾(6)、兑(7)、艮(8)、离(9)，对应八方和五行 - **天盘/地盘**：天干在九宫中的分布，天盘为动态，地盘为静态 - **九星**：天蓬、天芮、天冲、天辅、天禽、天心、天柱、天任、天英 - **八门**：休门、生门、伤门、杜门、景门、死门、惊门、开门 - **八神**：值符、腾蛇、太阴、六合、白虎、玄武、九地、九天 - **值符/值使**：当前时刻的主星和主门，是判断吉凶的核心 ## 解读要点 1. **看值符值使**：值符星和值使门的状态决定总体吉凶 2. **看用神宫位**：根据用途找到对应宫位（如求财看生门所在宫） 3. **看三奇六仪**：乙丙丁为三奇（吉），戊己庚辛壬癸为六仪 4. **看格局**：天盘与地盘天干组合形成吉格（如龙遁、虎遁）或凶格（如击刑、入墓） 5. **看空亡**：落入空亡的宫位力量减弱 ## 常见用途 - 求财：重点看生门、戊 - 事业：重点看开门、值使 - 感情：重点看六合、天后 - 出行：重点看三奇得使、门迫否

Retrieves and queries up-to-date documentation and code examples from Context7 for any programming library or framework. You must call 'resolve-library-id' first to obtain the exact Context7-compatible library ID required to use this tool, UNLESS the user explicitly provides a library ID in the format '/org/project' or '/org/project/version' in their query. IMPORTANT: Do not call this tool more than 3 times per question. If you cannot find what you need after 3 calls, use the best information you have.

Run a read-only shell-like query against a virtualized, in-memory filesystem rooted at `/` that contains ONLY the Cerebrium documentation pages and OpenAPI specs. This is NOT a shell on any real machine — nothing runs on the user's computer, the server host, or any network. The filesystem is a sandbox backed by documentation chunks. This is how you read documentation pages: there is no separate "get page" tool. To read a page, pass its `.mdx` path (e.g. `/quickstart.mdx`, `/api-reference/create-customer.mdx`) to `head` or `cat`. To search the docs with exact keyword or regex matches, use `rg`. To understand the docs structure, use `tree` or `ls`. **Workflow:** Start with the search tool for broad or conceptual queries like "how to authenticate" or "rate limiting". Use this tool when you need exact keyword/regex matching, structural exploration, or to read the full content of a specific page by path. Supported commands: rg (ripgrep), grep, find, tree, ls, cat, head, tail, stat, wc, sort, uniq, cut, sed, awk, jq, plus basic text utilities. No writes, no network, no process control. Run `--help` on any command for usage. Each call is STATELESS: the working directory always resets to `/` and no shell variables, aliases, or history carry over between calls. If you need to operate in a subdirectory, chain commands in one call with `&&` or pass absolute paths (e.g., `cd /api-reference && ls` or `ls /api-reference`). Do NOT assume that `cd` in one call affects the next call. Examples: - `tree / -L 2` — see the top-level directory layout - `rg -il "rate limit" /` — find all files mentioning "rate limit" - `rg -C 3 "apiKey" /api-reference/` — show matches with 3 lines of context around each hit - `head -80 /quickstart.mdx` — read the top 80 lines of a specific page - `head -80 /quickstart.mdx /installation.mdx /guides/first-deploy.mdx` — read multiple pages in one call - `cat /api-reference/create-customer.mdx` — read a full page when you need everything - `cat /openapi/spec.json | jq '.paths | keys'` — list OpenAPI endpoints Output is truncated to 30KB per call. Prefer targeted `rg -C` or `head -N` over broad `cat` on large files. To read only the relevant sections of a large file, use `rg -C 3 "pattern" /path/file.mdx`. Batch multiple file reads into a single `head` or `cat` call whenever possible. When referencing pages in your response to the user, convert filesystem paths to URL paths by removing the `.mdx` extension. For example, `/quickstart.mdx` becomes `/quickstart` and `/api-reference/overview.mdx` becomes `/api-reference/overview`.

Run a read-only shell-like query against a virtualized, in-memory filesystem rooted at `/` that contains ONLY the Cerebrium documentation pages and OpenAPI specs. This is NOT a shell on any real machine — nothing runs on the user's computer, the server host, or any network. The filesystem is a sandbox backed by documentation chunks. This is how you read documentation pages: there is no separate "get page" tool. To read a page, pass its `.mdx` path (e.g. `/quickstart.mdx`, `/api-reference/create-customer.mdx`) to `head` or `cat`. To search the docs with exact keyword or regex matches, use `rg`. To understand the docs structure, use `tree` or `ls`. **Workflow:** Start with the search tool for broad or conceptual queries like "how to authenticate" or "rate limiting". Use this tool when you need exact keyword/regex matching, structural exploration, or to read the full content of a specific page by path. Supported commands: rg (ripgrep), grep, find, tree, ls, cat, head, tail, stat, wc, sort, uniq, cut, sed, awk, jq, plus basic text utilities. No writes, no network, no process control. Run `--help` on any command for usage. Each call is STATELESS: the working directory always resets to `/` and no shell variables, aliases, or history carry over between calls. If you need to operate in a subdirectory, chain commands in one call with `&&` or pass absolute paths (e.g., `cd /api-reference && ls` or `ls /api-reference`). Do NOT assume that `cd` in one call affects the next call. Examples: - `tree / -L 2` — see the top-level directory layout - `rg -il "rate limit" /` — find all files mentioning "rate limit" - `rg -C 3 "apiKey" /api-reference/` — show matches with 3 lines of context around each hit - `head -80 /quickstart.mdx` — read the top 80 lines of a specific page - `head -80 /quickstart.mdx /installation.mdx /guides/first-deploy.mdx` — read multiple pages in one call - `cat /api-reference/create-customer.mdx` — read a full page when you need everything - `cat /openapi/spec.json | jq '.paths | keys'` — list OpenAPI endpoints Output is truncated to 30KB per call. Prefer targeted `rg -C` or `head -N` over broad `cat` on large files. To read only the relevant sections of a large file, use `rg -C 3 "pattern" /path/file.mdx`. Batch multiple file reads into a single `head` or `cat` call whenever possible. When referencing pages in your response to the user, convert filesystem paths to URL paths by removing the `.mdx` extension. For example, `/quickstart.mdx` becomes `/quickstart` and `/api-reference/overview.mdx` becomes `/api-reference/overview`.

Run a read-only shell-like query against a virtualized, in-memory filesystem rooted at `/` that contains ONLY the Cerebrium documentation pages and OpenAPI specs. This is NOT a shell on any real machine — nothing runs on the user's computer, the server host, or any network. The filesystem is a sandbox backed by documentation chunks. This is how you read documentation pages: there is no separate "get page" tool. To read a page, pass its `.mdx` path (e.g. `/quickstart.mdx`, `/api-reference/create-customer.mdx`) to `head` or `cat`. To search the docs with exact keyword or regex matches, use `rg`. To understand the docs structure, use `tree` or `ls`. **Workflow:** Start with the search tool for broad or conceptual queries like "how to authenticate" or "rate limiting". Use this tool when you need exact keyword/regex matching, structural exploration, or to read the full content of a specific page by path. Supported commands: rg (ripgrep), grep, find, tree, ls, cat, head, tail, stat, wc, sort, uniq, cut, sed, awk, jq, plus basic text utilities. No writes, no network, no process control. Run `--help` on any command for usage. Each call is STATELESS: the working directory always resets to `/` and no shell variables, aliases, or history carry over between calls. If you need to operate in a subdirectory, chain commands in one call with `&&` or pass absolute paths (e.g., `cd /api-reference && ls` or `ls /api-reference`). Do NOT assume that `cd` in one call affects the next call. Examples: - `tree / -L 2` — see the top-level directory layout - `rg -il "rate limit" /` — find all files mentioning "rate limit" - `rg -C 3 "apiKey" /api-reference/` — show matches with 3 lines of context around each hit - `head -80 /quickstart.mdx` — read the top 80 lines of a specific page - `head -80 /quickstart.mdx /installation.mdx /guides/first-deploy.mdx` — read multiple pages in one call - `cat /api-reference/create-customer.mdx` — read a full page when you need everything - `cat /openapi/spec.json | jq '.paths | keys'` — list OpenAPI endpoints Output is truncated to 30KB per call. Prefer targeted `rg -C` or `head -N` over broad `cat` on large files. To read only the relevant sections of a large file, use `rg -C 3 "pattern" /path/file.mdx`. Batch multiple file reads into a single `head` or `cat` call whenever possible. When referencing pages in your response to the user, convert filesystem paths to URL paths by removing the `.mdx` extension. For example, `/quickstart.mdx` becomes `/quickstart` and `/api-reference/overview.mdx` becomes `/api-reference/overview`.

Run a read-only shell-like query against a virtualized, in-memory filesystem rooted at `/` that contains ONLY the Cerebrium documentation pages and OpenAPI specs. This is NOT a shell on any real machine — nothing runs on the user's computer, the server host, or any network. The filesystem is a sandbox backed by documentation chunks. This is how you read documentation pages: there is no separate "get page" tool. To read a page, pass its `.mdx` path (e.g. `/quickstart.mdx`, `/api-reference/create-customer.mdx`) to `head` or `cat`. To search the docs with exact keyword or regex matches, use `rg`. To understand the docs structure, use `tree` or `ls`. **Workflow:** Start with the search tool for broad or conceptual queries like "how to authenticate" or "rate limiting". Use this tool when you need exact keyword/regex matching, structural exploration, or to read the full content of a specific page by path. Supported commands: rg (ripgrep), grep, find, tree, ls, cat, head, tail, stat, wc, sort, uniq, cut, sed, awk, jq, plus basic text utilities. No writes, no network, no process control. Run `--help` on any command for usage. Each call is STATELESS: the working directory always resets to `/` and no shell variables, aliases, or history carry over between calls. If you need to operate in a subdirectory, chain commands in one call with `&&` or pass absolute paths (e.g., `cd /api-reference && ls` or `ls /api-reference`). Do NOT assume that `cd` in one call affects the next call. Examples: - `tree / -L 2` — see the top-level directory layout - `rg -il "rate limit" /` — find all files mentioning "rate limit" - `rg -C 3 "apiKey" /api-reference/` — show matches with 3 lines of context around each hit - `head -80 /quickstart.mdx` — read the top 80 lines of a specific page - `head -80 /quickstart.mdx /installation.mdx /guides/first-deploy.mdx` — read multiple pages in one call - `cat /api-reference/create-customer.mdx` — read a full page when you need everything - `cat /openapi/spec.json | jq '.paths | keys'` — list OpenAPI endpoints Output is truncated to 30KB per call. Prefer targeted `rg -C` or `head -N` over broad `cat` on large files. To read only the relevant sections of a large file, use `rg -C 3 "pattern" /path/file.mdx`. Batch multiple file reads into a single `head` or `cat` call whenever possible. When referencing pages in your response to the user, convert filesystem paths to URL paths by removing the `.mdx` extension. For example, `/quickstart.mdx` becomes `/quickstart` and `/api-reference/overview.mdx` becomes `/api-reference/overview`.

Manage per-connection SQL rules: block dangerous patterns, require WHERE on large tables, log PII access. [ARCHITECT tier]

Return recent queries executed through ThinAir with timing, row counts, and status. [BUILD tier]

Execute a read-only SQL query against the target connection. ONLY SELECT / WITH / EXPLAIN permitted. Write dialect-appropriate SQL for the connection's engine — use PostgreSQL syntax for postgres connections (`SELECT NOW()`, `LIMIT`, `ILIKE`), T-SQL for mssql (`SELECT GETDATE()`, `TOP N`, `LIKE`), MySQL for mysql (`SELECT NOW()`, `LIMIT`). Response meta includes `connection` + `dialect` so you know which syntax worked; reuse that dialect in follow-up calls. Default LIMIT 100 unless the user asks for all rows.

Gets a list of pages contained in a database, filtered and ordered by the provided criteria.

Check current API usage, daily limit, plan name, and upgrade options.

Assess a request for risk using the Radar engine and receive a verdict.