{"slug": "28-mcp-tools-in-one-connection-a-developer-s-field-guide", "title": "28 MCP Tools in One Connection: A Developer's Field Guide", "summary": "Spiderbrain's MCP server exposes 28 tools for codebase work, organized into structural queries, memory tree access, and blueprint operations. The developer emphasizes a 'moment-based' dispatch pattern, where tools are called based on workflow situations like new project or refactoring, rather than data model operations. This approach, they argue, is more effective than having many tools without clear dispatch logic.", "body_md": "Model Context Protocol (MCP) lets an AI agent call structured tools instead of guessing from raw text dumped into a context window. Fine in theory — but most MCP integrations just expose \"read file\" and \"search,\" which barely improves on copy-pasting code into a chat window.\n\nThe real question: what tools should an agent actually have access to, for codebase work specifically?\n\nSpiderbrain's MCP server exposes 28 tools, split into three functional groups:\n\nStructural queries a flat context window literally cannot answer:\n\n```\nblast_radius(path)      // reach, callers, transitive impact\nscore_summary()          // keystones, blindspots, node scores\ncommunities()            // clusters, boundary edges, paths\ninsights(area)           // summaries, diffs, navigation, search\n```\n\nAccess to the Memory Tree — a project's accumulated decisions and constraints, persisted across sessions:\n\n```\nmemory_recall(files)     // scoped brief for current context\nmemory_search(query)     // semantic search over the full tree\nmemory_add(node)         // write with explicit user ratification\nmemory_browse(scope)     // browse by scope or recency\nspiderbrain_blueprint_read()   // current nodes, edges, annotations\nspiderbrain_blueprint_draw()   // AI proposes canvas edits — requires human acceptance\n```\n\nDon't call all 28 on every turn — that's noise, not context. Here's the practical dispatch table:\n\n| Situation | Call |\n|---|---|\n| New project |\n`score_summary()` + `top_spikescore(10)`\n|\n| Before editing a file |\n`node(path)` + `blast_radius(path)`\n|\n| Session start | `memory_recall(files)` |\n| Planning a refactor |\n`communities()` + `boundary_edges()`\n|\n| Onboarding to unfamiliar code |\n`blueprint_read()` + `insights(area)`\n|\n\nThis mapping matters more than tool count. An agent with 28 tools and no dispatch logic will either call none of them or call all of them — both are worse than an agent with 5 tools it actually reaches for at the right moment.\n\nThe pattern that generalizes beyond this specific server: **structure your MCP tools around the moments in a workflow, not around your data model.** \"New project,\" \"before editing,\" \"session start,\" \"refactor,\" and \"onboarding\" are moments a developer or agent actually hits. \"Get node,\" \"get edge,\" \"get community\" are just database operations wearing a tool-call costume.\n\nLink : [https://spiderbrain.ai/](https://spiderbrain.ai/)\n\nLinkedin : [https://www.linkedin.com/company/spiderbrain](https://www.linkedin.com/company/spiderbrain)\n\nDiscuss: has anyone else settled on a \"moment-based\" tool dispatch pattern for their own MCP servers? Curious what situations you gate on.", "url": "https://wpnews.pro/news/28-mcp-tools-in-one-connection-a-developer-s-field-guide", "canonical_source": "https://dev.to/neha_jakate_/28-mcp-tools-in-one-connection-a-developers-field-guide-3m6e", "published_at": "2026-08-02 10:55:16+00:00", "updated_at": "2026-08-02 11:15:03.253139+00:00", "lang": "en", "topics": ["ai-agents", "developer-tools", "ai-infrastructure"], "entities": ["Spiderbrain", "MCP"], "alternates": {"html": "https://wpnews.pro/news/28-mcp-tools-in-one-connection-a-developer-s-field-guide", "markdown": "https://wpnews.pro/news/28-mcp-tools-in-one-connection-a-developer-s-field-guide.md", "text": "https://wpnews.pro/news/28-mcp-tools-in-one-connection-a-developer-s-field-guide.txt", "jsonld": "https://wpnews.pro/news/28-mcp-tools-in-one-connection-a-developer-s-field-guide.jsonld"}}