Skip to content

Conversation

mattjoyce
Copy link

Add mcp-construe to Community Servers

Description

Adding mcp-construe to the community servers section - a FastMCP server for intelligent Obsidian vault context management with frontmatter filtering and secure bidirectional knowledge operations.

Server Details

Motivation and Context

mcp-construe solves a critical problem for AI-powered knowledge work: intelligently curating personal knowledge from Obsidian vaults for AI conversations.

Key problems solved:

  • Context Overload: Automatically filters and chunks large knowledge bases to stay within context limits (95k chars/chunk)
  • Smart Discovery: Flexible frontmatter-based filtering enables precise context loading by tags, properties, and content patterns
  • Bidirectional Knowledge: Read from vault for context AND write back discoveries with proper frontmatter
  • Security: Robust protection system prevents unauthorized modifications to personal knowledge base

This enables users to:

  1. Load exactly the right context for each AI conversation
  2. Build up curated knowledge bases that AI can navigate intelligently
  3. Capture AI discoveries back to their vault safely
  4. Progressive disclosure - browse metadata before loading full content

How Has This Been Tested?

Extensively tested with Claude Desktop and Claude Code:

Scenarios tested:

  • Context Loading: Fetching files by context type (personal, work, projects) with automatic chunking
  • Smart Filtering: Complex frontmatter queries with AND/OR tag logic across 1,200+ markdown files
  • Content Search: Regex-based vault search with context snippets and pagination
  • File Creation: Protected file writing with frontmatter validation
  • Chunking: Large result sets (>95k chars) properly split across multiple chunks
  • Progressive Disclosure: Frontmatter index browsing before loading full content

Production use:

  • Powers the PAI (Personal AI Infrastructure) system
  • Manages 1,200+ markdown files in Obsidian vault
  • Daily use for context-aware AI conversations
  • Integrated with healthcare AI workflow and conservation technology projects

Breaking Changes

No breaking changes - this is a new server addition to the community servers list.

Types of changes

  • Bug fix (non-breaking change which fixes an issue)
  • New feature (non-breaking change which adds functionality)
  • Breaking change (fix or feature that would cause existing functionality to change)
  • Documentation update

Checklist

  • I have read the MCP Protocol Documentation
  • My changes follows MCP security best practices
  • I have updated the server's README accordingly
  • I have tested this with an LLM client
  • My code follows the repository's style guidelines
  • New and existing tests pass locally
  • I have added appropriate error handling
  • I have documented all environment variables and configuration options

Additional context

Security Features:

  • Dual-layer protection: validates both new files AND existing files during overwrites
  • Configurable protection properties (author, context, owner)
  • Prevents unauthorized vault modifications
  • Safe YAML frontmatter parsing and generation

Key Tools Provided:

  • fetch_context() - Load files by context type
  • fetch_matching_files() - Flexible property/tag filtering
  • fetch_frontmatter_index() - Lightweight metadata browsing
  • search_vault_content() - Content-based discovery with regex
  • create_file() - Protected file creation with frontmatter
  • Plus chunking management tools for large result sets

Real-world Impact:
Used daily in healthcare AI innovation work and conservation technology projects, demonstrating practical value for knowledge workers using AI assistants with personal knowledge bases.

Two FastMCP servers for knowledge management and professional development:

- Construe: Intelligent Obsidian vault context management with frontmatter filtering
- Persona Sessions: Structured AI sessions for interview prep, reflection, and coaching

Both servers are production-tested and actively maintained.
Copy link
Member

@olaservo olaservo left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Are these official integrations or community servers? (It looks like these got added as official integrations.)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants