{ "data": { "category": { "area": "developer", "capabilities": [ "code.repo", "code.git", "code.docs", "code.deploy", "docs.microsoft" ], "description": "Repository platforms, version control, and up-to-date library documentation delivered straight into the context window.", "indexed": [ { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/braincloudservers-braincloud-mcp-helper.json", "kind": "mcp", "name": "brainCloud MCP Helper", "slug": "braincloudservers-braincloud-mcp-helper", "url": "https://www.anchorterminal.com/tools/braincloudservers-braincloud-mcp-helper" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/docfork-mcp.json", "kind": "mcp", "name": "docfork-mcp", "slug": "docfork-mcp", "url": "https://www.anchorterminal.com/tools/docfork-mcp" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/driflyte-mcp-server.json", "kind": "mcp", "name": "driflyte-mcp-server", "slug": "driflyte-mcp-server", "url": "https://www.anchorterminal.com/tools/driflyte-mcp-server" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/gitdiagram.json", "kind": "mcp", "name": "GitDiagram", "slug": "gitdiagram", "url": "https://www.anchorterminal.com/tools/gitdiagram" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/kinlab-kin.json", "kind": "mcp", "name": "Kin", "slug": "kinlab-kin", "url": "https://www.anchorterminal.com/tools/kinlab-kin" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/lastping.json", "kind": "mcp", "name": "LastPing", "slug": "lastping", "url": "https://www.anchorterminal.com/tools/lastping" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/lastseen-mortality.json", "kind": "mcp", "name": "lastseen mortality", "slug": "lastseen-mortality", "url": "https://www.anchorterminal.com/tools/lastseen-mortality" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/openlyuseful-linear-project.json", "kind": "mcp", "name": "Linear Project", "slug": "openlyuseful-linear-project", "url": "https://www.anchorterminal.com/tools/openlyuseful-linear-project" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/opencovenant-mizuki.json", "kind": "mcp", "name": "Mizuki", "slug": "opencovenant-mizuki", "url": "https://www.anchorterminal.com/tools/opencovenant-mizuki" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/pkgproof.json", "kind": "mcp", "name": "pkgproof", "slug": "pkgproof", "url": "https://www.anchorterminal.com/tools/pkgproof" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/previewdrop-cli.json", "kind": "mcp", "name": "PreviewDrop", "slug": "previewdrop-cli", "url": "https://www.anchorterminal.com/tools/previewdrop-cli" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/repowise.json", "kind": "mcp", "name": "Repowise", "slug": "repowise", "url": "https://www.anchorterminal.com/tools/repowise" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/eclipsesource-review-guard.json", "kind": "mcp", "name": "ReviewGuard", "slug": "eclipsesource-review-guard", "url": "https://www.anchorterminal.com/tools/eclipsesource-review-guard" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/sexai-mcp.json", "kind": "mcp", "name": "sexai-mcp", "slug": "sexai-mcp", "url": "https://www.anchorterminal.com/tools/sexai-mcp" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/sigit-cli.json", "kind": "mcp", "name": "siGit CLI", "slug": "sigit-cli", "url": "https://www.anchorterminal.com/tools/sigit-cli" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/social-superpowers.json", "kind": "mcp", "name": "Superpowers.social", "slug": "social-superpowers", "url": "https://www.anchorterminal.com/tools/social-superpowers" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/uploads-mcp.json", "kind": "mcp", "name": "uploads.sh", "slug": "uploads-mcp", "url": "https://www.anchorterminal.com/tools/uploads-mcp" } ], "indexedCount": 17, "json": "https://www.anchorterminal.com/categories/code.json", "name": "Code \u0026 developer platforms", "slug": "code", "test": "", "title": "Code, repository and documentation tools for AI agents", "toolCount": 8, "tools": [ "context7", "github-mcp-server", "azure-devops-mcp", "salesforce-dx-mcp", "ref", "sourcegraph-mcp", "git-reference-server", "microsoft-learn-mcp" ], "url": "https://www.anchorterminal.com/categories/code" }, "faq": [ { "answer": "Context7 has the highest benchmark score of the 8 ranked code, repository and documentation tools, 73 (BB). GitHub MCP Server is second with 70.3 (BB).", "question": "What are the highest-rated code, repository and documentation tools for AI agents?" }, { "answer": "2 of the 8 ranked here grade BB or better, the bar for agent-ready on the Anchor benchmark.", "question": "How many code, repository and documentation tools are agent-ready?" }, { "answer": "None of the ranked listings here accepts x402 for its main call yet.", "question": "Which code, repository and documentation tools accept x402 payments?" }, { "answer": "By the Anchor benchmark score out of 100, a weighted mean of the scored categories minus deductions for negative events, from public evidence re-checked as vendors change. Listings cannot pay for a place. The latest assessment behind this page is from 8 October 2026.", "question": "How is this list ranked?" } ], "howToChoose": [ { "label": "Read and write tool scope", "detail": "Check which tools can read and which can write to repositories, and whether you can filter them, since an agent needs only the calls its task requires." }, { "label": "Private repository access", "detail": "Check how the server authenticates, which access controls apply and whether private sources stay private, since a broad token lets an agent read code it should not see." }, { "label": "Result limits and metered credits", "detail": "Check result limits and which tools draw on metered credits, since a documentation lookup or repository search can use up a plan's allowance inside a single agent run." }, { "label": "Documentation version and freshness", "detail": "Check how library documentation is versioned and how often it updates, since an agent generating code against an outdated API version may produce calls that fail." } ], "picks": [ { "also": { "name": "GitHub MCP Server", "slug": "github-mcp-server", "why": "BB, 70.3/100" }, "name": "Context7", "need": "Highest score overall", "slug": "context7", "why": "BB, 73/100 on the benchmark" }, { "name": "Azure DevOps MCP Server", "need": "Reliability", "slug": "azure-devops-mcp", "why": "72/100 on reliability, against 68 for the overall leader" }, { "name": "GitHub MCP Server", "need": "Agent ergonomics", "slug": "github-mcp-server", "why": "78/100 on agent ergonomics, against 69 for the overall leader" }, { "name": "GitHub MCP Server", "need": "Security \u0026 auth", "slug": "github-mcp-server", "why": "84/100 on security \u0026 auth, against 69 for the overall leader" }, { "name": "GitHub MCP Server", "need": "Transparency \u0026 trust", "slug": "github-mcp-server", "why": "83/100 on transparency \u0026 trust, against 72 for the overall leader" }, { "name": "Sourcegraph MCP Server", "need": "Self-hosting under an open licence", "slug": "sourcegraph-mcp", "why": "self-hosted, Proprietary licence" } ], "ranked": 8, "shortlist": [ { "bestFor": "Coding agents that need current library or framework docs and examples on demand, at little or no cost.", "grade": "BB", "name": "Context7", "position": 1, "price": "$5 / 1k calls", "score": 73, "slug": "context7", "strengths": [ "Two tools, both annotated `readOnlyHint: true` and `idempotentHint: true`", "Anonymous calls work on https://mcp.context7.com/mcp, and the free plan allows 1,000 calls a month with no card", "429s carry `Retry-After` and `RateLimit-*` headers, documented with backoff guidance" ], "url": "https://www.anchorterminal.com/tools/context7", "verdict": "Two tools, both annotated `readOnlyHint: true` and `idempotentHint: true`. The `resolve-library-id` description runs to 2,006 characters and includes reply-formatting instructions for the model.", "weaknesses": [ "The `resolve-library-id` description runs to 2,006 characters and includes reply-formatting instructions for the model", "No page, limit or token parameter since 2.0.0, so response size depends on how narrow the query is", "The 11 August 2026 mcp.context7.com outage (503s from a stream leak in 4.0.0) appears in the changelog but not on status.upstash.com" ], "where": "both", "x402": "no" }, { "bestFor": "Agents that read and change GitHub repositories, issues, pull requests and Actions, from triage bots to coding agents.", "grade": "BB", "name": "GitHub MCP Server", "position": 2, "price": "Free · OSS", "score": 70.3, "slug": "github-mcp-server", "strengths": [ "OAuth with scopes by default, fine-grained PATs and GitHub App tokens for headless runs, and per-call scope challenges since v1.11.0", "Every remote toolset has a `/readonly` URL, and `--read-only` drops write tools even when named in `--tools`", "`delete_repository` needs the user to type the full repository name through elicitation" ], "url": "https://www.anchorterminal.com/tools/github-mcp-server", "verdict": "OAuth with scopes by default, fine-grained PATs and GitHub App tokens for headless runs, and per-call scope challenges since v1.11.0. 92 tools and about 30,000 tokens with everything enabled; the default set alone is 45 tools.", "weaknesses": [ "92 tools and about 30,000 tokens with everything enabled; the default set alone is 45 tools", "27 of 35 write tools leave `destructiveHint` unset, and read tools don't set `idempotentHint` (issue #3281 is open)", "Two advisories in 2026, one of them cross-user GraphQL client confusion in HTTP mode, both fixed" ], "where": "both", "x402": "no" }, { "bestFor": "Agents working in an Azure DevOps organisation on work items, pull requests, pipelines and test plans, most easily from Microsoft's own clients.", "grade": "B", "name": "Azure DevOps MCP Server", "position": 3, "price": "$6 / seat-mo", "score": 66.3, "slug": "azure-devops-mcp", "strengths": [ "`X-MCP-Readonly`, `X-MCP-Toolsets` and `X-MCP-Tools` headers trim the remote server to read tools, eight toolsets or named tools", "Remote access uses Microsoft Entra OAuth with Azure DevOps scopes, and a call needs both the app scope and the user's own permission", "All 45 tools in the local package carry read-only, destructive, idempotent and open-world hints, checked by a test" ], "url": "https://www.anchorterminal.com/tools/azure-devops-mcp", "verdict": "The hosted endpoint takes Entra OAuth with Azure DevOps scopes, an `X-MCP-Readonly` header and toolset or per-tool filters, and the MIT local package annotates every tool. The remote server is still labelled public preview, and Claude Code and Cursor need a custom Entra app registration with admin consent before they can use it.", "weaknesses": [ "The remote server was announced as a public preview on 17 March 2026 and no general availability notice was found", "Claude Code and Cursor need a custom Entra app registration with admin consent, and Claude Desktop and Codex can't use the remote server", "Version 2.9.0 renamed the local tools in a minor release on 29 July 2026, with the README warning added two days later" ], "where": "both", "x402": "no" }, { "bestFor": "Coding agents building and deploying on the Salesforce platform (metadata, Apex tests, LWC guidance, `DevOps Center`).", "grade": "C", "name": "Salesforce DX MCP Server", "position": 4, "price": "Free · OSS", "score": 59.5, "slug": "salesforce-dx-mcp", "strengths": [ "Tools pass org usernames, never tokens, and `--orgs` limits which authorised orgs the server can reach", "Toolsets, `--tools` and NON-GA gating keep the default surface smaller than the 88-tool total", "Unit tests on Linux and Windows plus end-to-end tests on every branch push" ], "url": "https://www.anchorterminal.com/tools/salesforce-dx-mcp", "verdict": "Tools pass org usernames, never tokens, and `--orgs` limits which authorised orgs the server can reach. 88 tools; Salesforce's README warns that enabling all of them overwhelms the context.", "weaknesses": [ "88 tools; Salesforce's README warns that enabling all of them overwhelms the context", "Ten `DevOps Center` tools and `delete_org` carry no annotations", "One release (0.30.15, 9 July 2026) in the last 90 days, and 2025 bug reports still open" ], "where": "local", "x402": "no" }, { "bestFor": "Coding agents that need current library documentation plus search over a team's private repositories and PDFs through two small tools.", "grade": "C", "name": "Ref", "position": 5, "price": "$10 / 1k calls", "score": 59.4, "slug": "ref", "strengths": [ "Two tools, `ref_search_documentation` and `ref_read_url`, each with one required string and `readOnlyHint: true`", "OAuth at `https://api.ref.tools/mcp` advertises two scopes, `public_docs:read` and `private_docs:read`", "The activity log at ref.tools/activity records user identity, tool calls and arguments, per the security page" ], "url": "https://www.anchorterminal.com/tools/ref", "verdict": "Two read-only tools with short definitions, OAuth with separate scopes for public and private documentation, and an activity log of every tool call. The API key may be passed in the URL query string, no rate limit or 429 guidance is documented, and the free allowance is 200 one-time credits.", "weaknesses": [ "The install docs give `?apiKey=YOUR_API_KEY` in the URL as a documented way to authenticate", "No rate limit, 429 handling or SLA was found in the reviewed documentation", "The free plan is 200 one-time credits, and an account needs a browser sign-up" ], "where": "both", "x402": "no" }, { "bestFor": "A company that already runs Sourcegraph and wants agents to search and navigate many repositories with compiler-accurate definitions and references.", "grade": "C", "name": "Sourcegraph MCP Server", "position": 6, "price": "Paid", "score": 57.5, "slug": "sourcegraph-mcp", "strengths": [ "All 16 documented tools read and none writes to a repository, with three endpoints so a client can load 9, 16 or 2 tools", "OAuth 2.0 with PKCE and dynamic client registration, registered clients held to the `mcp` scope, and access tokens that can carry the same scope", "Repository permissions apply to every call, and admins can gate MCP by role (`MCP#ACCESS`) or switch off single tools with `mcp.tools.disabled`" ], "url": "https://www.anchorterminal.com/tools/sourcegraph-mcp", "verdict": "Sixteen read-only tools search and navigate code across every repository an instance indexes, with OAuth limited to an `mcp` scope and repository permissions enforced on each call. Access needs an Enterprise contract, priced from $16,000 a year, and no request rate limits were found in the reviewed documentation.", "weaknesses": [ "Enterprise plans only, from a $16,000 minimum annual contract through sales. A Cloud trial is by request and subject to eligibility", "No request rate limits, 429 behaviour or tool error catalogue found in the reviewed documentation", "`code_finder` and Deep Search draw on the instance's credit entitlement and return an error once the quota is used up" ], "where": "local", "x402": "no" }, { "bestFor": "Local commit loops in a sandboxed checkout where an agent stages, commits and branches.", "grade": "D", "name": "Git (MCP reference server)", "position": 7, "price": "Free · OSS", "score": 51.9, "slug": "git-reference-server", "strengths": [ "Twelve tools of about 1,400 tokens in total, each with `readOnlyHint` or `destructiveHint` set", "`--repository` and MCP roots confine every call to allowed paths, with checks hardened after the December 2025 advisories", "Refs and paths starting with `-` are rejected, closing the argument-injection class" ], "url": "https://www.anchorterminal.com/tools/git-reference-server", "verdict": "Twelve tools of about 1,400 tokens in total, each with `readOnlyHint` or `destructiveHint` set. Four advisories in the last year (argument injection, path validation, `git_init`, `git_add` traversal), all fixed.", "weaknesses": [ "Four advisories in the last year (argument injection, path validation, `git_init`, `git_add` traversal), all fixed", "SECURITY.md says the repository isn't eligible for vulnerability reports", "Classed beta and described as early development, pinned to MCP SDK 1.x" ], "where": "local", "x402": "no" }, { "bestFor": "Coding and support agents working on Azure, .NET, Microsoft 365 or other Microsoft technologies, at no cost.", "grade": "D", "name": "Microsoft Learn MCP Server", "position": 8, "price": "Free", "score": 47.9, "slug": "microsoft-learn-mcp", "strengths": [ "No account, key or card, so an agent connects in one step", "Three tools, each a read with one required parameter", "`maxTokenBudget` on the URL caps search-result size" ], "url": "https://www.anchorterminal.com/tools/microsoft-learn-mcp", "verdict": "Access requires no account, key or card. No status page was found, and the FAQ does not specify numeric rate limits.", "weaknesses": [ "No status page, and the FAQ admits rate limits without giving numbers", "Microsoft says tools can change at any time and doesn't version them", "No statement of what the endpoint logs or keeps" ], "where": "hosted", "x402": "no" } ], "updated": "2026-10-08" }, "kind": "anchor.page", "links": { "api": "https://www.anchorterminal.com/api/v1/index.json", "html": "https://www.anchorterminal.com/best/code/", "json": "https://www.anchorterminal.com/best/code/index.json", "llms": "https://www.anchorterminal.com/llms.txt", "markdown": "https://www.anchorterminal.com/best/code/index.md", "slim": "https://www.anchorterminal.com/best/code/index.min.md" }, "markdown": "All 8 ranked code, repository and documentation tools on the Anchor benchmark, with a pick for each need and where each one falls short. Scores come from public evidence, re-checked as vendors change.\n\n- Ranked: 8 · agent-ready (BB or better): 2 · accept x402: 0 · hosted endpoints: 5\n- Full ranked table: https://www.anchorterminal.com/categories/code.md\n- Head-to-head comparisons: https://www.anchorterminal.com/compare/code/index.md (8)\n- Methodology: https://www.anchorterminal.com/benchmark/index.md\n\n## The shortlist\n\n| # | Tool | Grade | Score | Best for | Price | Where |\n| --- | --- | --- | --- | --- | --- | --- |\n| 1 | [Context7](https://www.anchorterminal.com/tools/context7.md) | BB | 73 | Coding agents that need current library or framework docs and examples on demand, at little or no cost. | $5 / 1k calls | hosted and local |\n| 2 | [GitHub MCP Server](https://www.anchorterminal.com/tools/github-mcp-server.md) | BB | 70.3 | Agents that read and change GitHub repositories, issues, pull requests and Actions, from triage bots to coding agents. | Free · OSS | hosted and local |\n| 3 | [Azure DevOps MCP Server](https://www.anchorterminal.com/tools/azure-devops-mcp.md) | B | 66.3 | Agents working in an Azure DevOps organisation on work items, pull requests, pipelines and test plans, most easily from Microsoft's own clients. | $6 / seat-mo | hosted and local |\n| 4 | [Salesforce DX MCP Server](https://www.anchorterminal.com/tools/salesforce-dx-mcp.md) | C | 59.5 | Coding agents building and deploying on the Salesforce platform (metadata, Apex tests, LWC guidance, \\`DevOps Center\\`). | Free · OSS | local |\n| 5 | [Ref](https://www.anchorterminal.com/tools/ref.md) | C | 59.4 | Coding agents that need current library documentation plus search over a team's private repositories and PDFs through two small tools. | $10 / 1k calls | hosted and local |\n| 6 | [Sourcegraph MCP Server](https://www.anchorterminal.com/tools/sourcegraph-mcp.md) | C | 57.5 | A company that already runs Sourcegraph and wants agents to search and navigate many repositories with compiler-accurate definitions and references. | Paid | local |\n| 7 | [Git (MCP reference server)](https://www.anchorterminal.com/tools/git-reference-server.md) | D | 51.9 | Local commit loops in a sandboxed checkout where an agent stages, commits and branches. | Free · OSS | local |\n| 8 | [Microsoft Learn MCP Server](https://www.anchorterminal.com/tools/microsoft-learn-mcp.md) | D | 47.9 | Coding and support agents working on Azure, .NET, Microsoft 365 or other Microsoft technologies, at no cost. | Free | hosted |\n\n## Picks by need\n\n- Highest score overall: [Context7](https://www.anchorterminal.com/tools/context7.md), BB, 73/100 on the benchmark. Also [GitHub MCP Server](https://www.anchorterminal.com/tools/github-mcp-server.md), BB, 70.3/100.\n- Reliability: [Azure DevOps MCP Server](https://www.anchorterminal.com/tools/azure-devops-mcp.md), 72/100 on reliability, against 68 for the overall leader.\n- Agent ergonomics: [GitHub MCP Server](https://www.anchorterminal.com/tools/github-mcp-server.md), 78/100 on agent ergonomics, against 69 for the overall leader.\n- Security \u0026 auth: [GitHub MCP Server](https://www.anchorterminal.com/tools/github-mcp-server.md), 84/100 on security \u0026 auth, against 69 for the overall leader.\n- Transparency \u0026 trust: [GitHub MCP Server](https://www.anchorterminal.com/tools/github-mcp-server.md), 83/100 on transparency \u0026 trust, against 72 for the overall leader.\n- Self-hosting under an open licence: [Sourcegraph MCP Server](https://www.anchorterminal.com/tools/sourcegraph-mcp.md), self-hosted, Proprietary licence.\n\n## How to choose\n\n- Read and write tool scope: Check which tools can read and which can write to repositories, and whether you can filter them, since an agent needs only the calls its task requires.\n- Private repository access: Check how the server authenticates, which access controls apply and whether private sources stay private, since a broad token lets an agent read code it should not see.\n- Result limits and metered credits: Check result limits and which tools draw on metered credits, since a documentation lookup or repository search can use up a plan's allowance inside a single agent run.\n- Documentation version and freshness: Check how library documentation is versioned and how often it updates, since an agent generating code against an outdated API version may produce calls that fail.\n\n## Each one in detail\n\n### 1. Context7, BB 73/100\n\nServes up-to-date, version-specific library documentation and code examples into agent prompts via two tools (resolve-library-id, query-docs).\n\n- Verdict: Two tools, both annotated `readOnlyHint: true` and `idempotentHint: true`. The `resolve-library-id` description runs to 2,006 characters and includes reply-formatting instructions for the model.\n- Choose it for: Coding agents that need current library or framework docs and examples on demand, at little or no cost.\n- Strength: Two tools, both annotated `readOnlyHint: true` and `idempotentHint: true`\n- Strength: Anonymous calls work on https://mcp.context7.com/mcp, and the free plan allows 1,000 calls a month with no card\n- Strength: 429s carry `Retry-After` and `RateLimit-*` headers, documented with backoff guidance\n- Weakness: The `resolve-library-id` description runs to 2,006 characters and includes reply-formatting instructions for the model\n- Weakness: No page, limit or token parameter since 2.0.0, so response size depends on how narrow the query is\n- Weakness: The 11 August 2026 mcp.context7.com outage (503s from a stream leak in 4.0.0) appears in the changelog but not on status.upstash.com\n- Price: $5 / 1k calls · Auth: OAuth or key · x402: no · Where: hosted and local\n- Full assessment: https://www.anchorterminal.com/tools/context7.md\n\n### 2. GitHub MCP Server, BB 70.3/100\n\nGitHub's official MCP server (Go) exposing repositories, issues, pull requests, Actions, code security, discussions, gists, notifications, projects and more as toolsets.\n\n- Verdict: OAuth with scopes by default, fine-grained PATs and GitHub App tokens for headless runs, and per-call scope challenges since v1.11.0. 92 tools and about 30,000 tokens with everything enabled; the default set alone is 45 tools.\n- Choose it for: Agents that read and change GitHub repositories, issues, pull requests and Actions, from triage bots to coding agents.\n- Strength: OAuth with scopes by default, fine-grained PATs and GitHub App tokens for headless runs, and per-call scope challenges since v1.11.0\n- Strength: Every remote toolset has a `/readonly` URL, and `--read-only` drops write tools even when named in `--tools`\n- Strength: `delete_repository` needs the user to type the full repository name through elicitation\n- Weakness: 92 tools and about 30,000 tokens with everything enabled; the default set alone is 45 tools\n- Weakness: 27 of 35 write tools leave `destructiveHint` unset, and read tools don't set `idempotentHint` (issue #3281 is open)\n- Weakness: Two advisories in 2026, one of them cross-user GraphQL client confusion in HTTP mode, both fixed\n- Price: Free · OSS · Auth: OAuth or key · x402: no · Where: hosted and local\n- Full assessment: https://www.anchorterminal.com/tools/github-mcp-server.md\n\n### 3. Azure DevOps MCP Server, B 66.3/100\n\nMicrosoft's official MCP server for Azure DevOps Services. It gives agents work items, repositories, pull requests, pipelines, wikis, test plans and Advanced Security alerts, through a hosted endpoint with Microsoft Entra sign-in or a local npm package.\n\n- Verdict: The hosted endpoint takes Entra OAuth with Azure DevOps scopes, an `X-MCP-Readonly` header and toolset or per-tool filters, and the MIT local package annotates every tool. The remote server is still labelled public preview, and Claude Code and Cursor need a custom Entra app registration with admin consent before they can use it.\n- Choose it for: Agents working in an Azure DevOps organisation on work items, pull requests, pipelines and test plans, most easily from Microsoft's own clients.\n- Strength: `X-MCP-Readonly`, `X-MCP-Toolsets` and `X-MCP-Tools` headers trim the remote server to read tools, eight toolsets or named tools\n- Strength: Remote access uses Microsoft Entra OAuth with Azure DevOps scopes, and a call needs both the app scope and the user's own permission\n- Strength: All 45 tools in the local package carry read-only, destructive, idempotent and open-world hints, checked by a test\n- Weakness: The remote server was announced as a public preview on 17 March 2026 and no general availability notice was found\n- Weakness: Claude Code and Cursor need a custom Entra app registration with admin consent, and Claude Desktop and Codex can't use the remote server\n- Weakness: Version 2.9.0 renamed the local tools in a minor release on 29 July 2026, with the README warning added two days later\n- Price: $6 / seat-mo · Auth: OAuth or key · x402: no · Where: hosted and local\n- Full assessment: https://www.anchorterminal.com/tools/azure-devops-mcp.md\n\n### 4. Salesforce DX MCP Server, C 59.5/100\n\nSalesforce's official local MCP server (`@salesforce/mcp`) for developing on the platform.\n\n- Verdict: Tools pass org usernames, never tokens, and `--orgs` limits which authorised orgs the server can reach. 88 tools; Salesforce's README warns that enabling all of them overwhelms the context.\n- Choose it for: Coding agents building and deploying on the Salesforce platform (metadata, Apex tests, LWC guidance, `DevOps Center`).\n- Strength: Tools pass org usernames, never tokens, and `--orgs` limits which authorised orgs the server can reach\n- Strength: Toolsets, `--tools` and NON-GA gating keep the default surface smaller than the 88-tool total\n- Strength: Unit tests on Linux and Windows plus end-to-end tests on every branch push\n- Weakness: 88 tools; Salesforce's README warns that enabling all of them overwhelms the context\n- Weakness: Ten `DevOps Center` tools and `delete_org` carry no annotations\n- Weakness: One release (0.30.15, 9 July 2026) in the last 90 days, and 2025 bug reports still open\n- Price: Free · OSS · Auth: OAuth or key · x402: no · Where: local\n- Full assessment: https://www.anchorterminal.com/tools/salesforce-dx-mcp.md\n\n### 5. Ref, C 59.4/100\n\nRef Context is a hosted MCP server from Ref Software, Inc. Its two tools search public documentation and a user's private repositories and PDFs, then read a page as Markdown trimmed to the relevant sections.\n\n- Verdict: Two read-only tools with short definitions, OAuth with separate scopes for public and private documentation, and an activity log of every tool call. The API key may be passed in the URL query string, no rate limit or 429 guidance is documented, and the free allowance is 200 one-time credits.\n- Choose it for: Coding agents that need current library documentation plus search over a team's private repositories and PDFs through two small tools.\n- Strength: Two tools, `ref_search_documentation` and `ref_read_url`, each with one required string and `readOnlyHint: true`\n- Strength: OAuth at `https://api.ref.tools/mcp` advertises two scopes, `public_docs:read` and `private_docs:read`\n- Strength: The activity log at ref.tools/activity records user identity, tool calls and arguments, per the security page\n- Weakness: The install docs give `?apiKey=YOUR_API_KEY` in the URL as a documented way to authenticate\n- Weakness: No rate limit, 429 handling or SLA was found in the reviewed documentation\n- Weakness: The free plan is 200 one-time credits, and an account needs a browser sign-up\n- Price: $10 / 1k calls · Auth: OAuth or key · x402: no · Where: hosted and local\n- Full assessment: https://www.anchorterminal.com/tools/ref.md\n- Against #1: https://www.anchorterminal.com/compare/context7-vs-ref.md\n\n### 6. Sourcegraph MCP Server, C 57.5/100\n\nSourcegraph's official MCP server gives AI agents code search, file reading, code navigation and commit and diff search across the repositories indexed by a company's Sourcegraph instance. It is built into Enterprise instances at /.api/mcp.\n\n- Verdict: Sixteen read-only tools search and navigate code across every repository an instance indexes, with OAuth limited to an `mcp` scope and repository permissions enforced on each call. Access needs an Enterprise contract, priced from $16,000 a year, and no request rate limits were found in the reviewed documentation.\n- Choose it for: A company that already runs Sourcegraph and wants agents to search and navigate many repositories with compiler-accurate definitions and references.\n- Strength: All 16 documented tools read and none writes to a repository, with three endpoints so a client can load 9, 16 or 2 tools\n- Strength: OAuth 2.0 with PKCE and dynamic client registration, registered clients held to the `mcp` scope, and access tokens that can carry the same scope\n- Strength: Repository permissions apply to every call, and admins can gate MCP by role (`MCP#ACCESS`) or switch off single tools with `mcp.tools.disabled`\n- Weakness: Enterprise plans only, from a $16,000 minimum annual contract through sales. A Cloud trial is by request and subject to eligibility\n- Weakness: No request rate limits, 429 behaviour or tool error catalogue found in the reviewed documentation\n- Weakness: `code_finder` and Deep Search draw on the instance's credit entitlement and return an error once the quota is used up\n- Price: Paid · Auth: OAuth or key · x402: no · Where: local\n- Full assessment: https://www.anchorterminal.com/tools/sourcegraph-mcp.md\n\n### 7. Git (MCP reference server), D 51.9/100\n\nPython reference server for reading and changing local Git repositories through twelve tools (status, staged and unstaged diffs, diff against a ref, add, commit, reset, log, create branch, checkout, show, branch list). `git_init` was removed in September 2025.\n\n- Verdict: Twelve tools of about 1,400 tokens in total, each with `readOnlyHint` or `destructiveHint` set. Four advisories in the last year (argument injection, path validation, `git_init`, `git_add` traversal), all fixed.\n- Choose it for: Local commit loops in a sandboxed checkout where an agent stages, commits and branches.\n- Strength: Twelve tools of about 1,400 tokens in total, each with `readOnlyHint` or `destructiveHint` set\n- Strength: `--repository` and MCP roots confine every call to allowed paths, with checks hardened after the December 2025 advisories\n- Strength: Refs and paths starting with `-` are rejected, closing the argument-injection class\n- Weakness: Four advisories in the last year (argument injection, path validation, `git_init`, `git_add` traversal), all fixed\n- Weakness: SECURITY.md says the repository isn't eligible for vulnerability reports\n- Weakness: Classed beta and described as early development, pinned to MCP SDK 1.x\n- Price: Free · OSS · Auth: None · x402: no · Where: local\n- Full assessment: https://www.anchorterminal.com/tools/git-reference-server.md\n- Disclosure: MCP started at Anthropic, which makes the Claude models our research agents and review panel run on (Anthropic donated it to the Agentic AI Foundation, a directed fund under the Linux Foundation, in December 2025), and this server is graded by the same checklist as every other listing.\n\n### 8. Microsoft Learn MCP Server, D 47.9/100\n\nMicrosoft's hosted documentation server at learn.microsoft.com/api/mcp.\n\n- Verdict: Access requires no account, key or card. No status page was found, and the FAQ does not specify numeric rate limits.\n- Choose it for: Coding and support agents working on Azure, .NET, Microsoft 365 or other Microsoft technologies, at no cost.\n- Strength: No account, key or card, so an agent connects in one step\n- Strength: Three tools, each a read with one required parameter\n- Strength: `maxTokenBudget` on the URL caps search-result size\n- Weakness: No status page, and the FAQ admits rate limits without giving numbers\n- Weakness: Microsoft says tools can change at any time and doesn't version them\n- Weakness: No statement of what the endpoint logs or keeps\n- Price: Free · Auth: None · x402: no · Where: hosted\n- Full assessment: https://www.anchorterminal.com/tools/microsoft-learn-mcp.md\n- Against #1: https://www.anchorterminal.com/compare/context7-vs-microsoft-learn-mcp.md\n\n## Head to head\n\n- [Context7 vs Ref](https://www.anchorterminal.com/compare/context7-vs-ref.md)\n- [Azure DevOps MCP Server vs GitHub MCP Server](https://www.anchorterminal.com/compare/azure-devops-mcp-vs-github-mcp-server.md)\n- [Azure DevOps MCP Server vs Salesforce DX MCP Server](https://www.anchorterminal.com/compare/azure-devops-mcp-vs-salesforce-dx-mcp.md)\n\n## Questions\n\n### What are the highest-rated code, repository and documentation tools for AI agents?\n\nContext7 has the highest benchmark score of the 8 ranked code, repository and documentation tools, 73 (BB). GitHub MCP Server is second with 70.3 (BB).\n\n### How many code, repository and documentation tools are agent-ready?\n\n2 of the 8 ranked here grade BB or better, the bar for agent-ready on the Anchor benchmark.\n\n### Which code, repository and documentation tools accept x402 payments?\n\nNone of the ranked listings here accepts x402 for its main call yet.\n\n### How is this list ranked?\n\nBy the Anchor benchmark score out of 100, a weighted mean of the scored categories minus deductions for negative events, from public evidence re-checked as vendors change. Listings cannot pay for a place. The latest assessment behind this page is from 8 October 2026.\n\n## How this list is made\n\nThe order is the Anchor benchmark score, the same number as on each listing. Each listing is graded from public evidence against the benchmark checklist, and the picks are worked out from those grades, prices and facts. No listing pays for its place, and paid audits or listing help never change a score.\n", "meta": { "attribution": "Anchor Terminal (https://www.anchorterminal.com)", "docs": "https://www.anchorterminal.com/docs/", "generatedAt": "2026-10-10", "license": "CC-BY-4.0", "method": "https://www.anchorterminal.com/benchmark/", "methodology": "0.4", "openapi": "https://www.anchorterminal.com/openapi.json", "preview": false, "run": "2026-10-01", "runLabel": "October 2026 research run" }, "page": { "breadcrumbs": [ { "name": "Home", "url": "https://www.anchorterminal.com/" }, { "name": "Best of", "url": "https://www.anchorterminal.com/best/" }, { "name": "Code \u0026 developer platforms", "url": "" } ], "description": "Context7 (BB), GitHub MCP Server (BB) and Azure DevOps MCP Server (B) lead the 8 ranked code, repository and documentation tools. Picks by need, strengths, weaknesses and prices from the Anchor benchmark.", "facts": [ "Context7 BB", "GitHub MCP Server BB", "Azure DevOps MCP Server B" ], "h1": "Best code, repository and documentation tools for AI agents", "image": "https://www.anchorterminal.com/assets/og/best-code.png", "path": "/best/code/", "published": "", "section": "tools", "title": "Best code, repository and documentation tools for AI agents in 2026", "toc": null, "updated": "2026-10-08", "url": "https://www.anchorterminal.com/best/code/" }, "tokens": { "markdown": 4650, "slim": 1230 }, "version": 1 }