Roadmap

Crow's development follows a milestone-based roadmap. Each milestone builds on the previous one, expanding the platform's project types, skills, and data backend integrations.

Milestone 1: Project Type System + Data Backend Registry (Current)

The foundation for everything that follows. The research server has been generalized into a project server that supports typed projects and pluggable data backends.

What shipped:

• Project types -- Projects now have a type field (research, data_connector) that determines their workflow
• Data backend registry -- Register external MCP servers as queryable data sources (crow_register_backend, crow_list_backends, crow_remove_backend, crow_backend_schema)
• data_backends table -- New database table for storing backend registrations
• Backward compatibility -- crow-research server name, /research/mcp endpoints, createResearchServer factory, and crow_research router category all continue to work as aliases
• Renamed server -- crow-projects is the canonical server name; crow_project_stats replaces crow_research_stats

Prerequisites for next milestones:

• Stable data backend protocol (register, query, capture)
• Project type system extensible for new verticals

Milestone 2: LMS Vertical

Bring learning management system data into Crow through purpose-built project types and backends.

Planned:

• lms_course project type -- Projects scoped to a specific course, with semester, section, and enrollment metadata
• Canvas LMS backend -- Pre-built data backend for Canvas (assignments, grades, modules, announcements, student submissions)
• LMS skill -- Behavioral prompt that teaches the AI how to work with course data, generate reports, and track student progress
• Grade analysis tools -- Tools for computing statistics, identifying at-risk students, and generating grade distribution reports
• Assignment workflow -- Create assignments, rubrics, and feedback drafts from project notes

Prerequisites:

• Milestone 1 complete
• Canvas MCP server operational (already listed as an external integration)

Milestone 3: Instructional Tools

Tools for course design, content creation, and pedagogical workflows.

Planned:

• curriculum project type -- Projects for designing course sequences, learning objectives, and assessment strategies
• Syllabus generator -- Skill that produces formatted syllabi from project notes and institutional templates
• Learning objective alignment -- Tools to map assignments to learning outcomes (Bloom's taxonomy tagging)
• Content scaffolding -- Break lecture topics into scaffolded learning sequences with suggested activities
• Assessment builder -- Generate quiz and exam items from source material, tagged by difficulty and objective

Prerequisites:

• Milestone 2 complete (LMS data informs curriculum decisions)
• Stable knowledge capture workflow from data backends

Milestone 4: Administrative and Institutional

Expand beyond individual courses to department-level and institutional workflows.

Planned:

• program_review project type -- Projects for accreditation, program assessment, and institutional reporting
• Multi-course analytics -- Aggregate data across courses for department-level insights (retention, DFW rates, enrollment trends)
• Accreditation evidence -- Tools to collect and organize evidence for accreditation standards (mapped to common frameworks like HLC, SACSCOC)
• Faculty workload -- Track teaching loads, committee assignments, and service obligations
• Institutional data backends -- Pre-built backends for SIS (Student Information System), HR, and financial systems

Prerequisites:

• Milestones 2 and 3 complete
• Institutional data access agreements and backend implementations
• Multi-user sharing workflows stable (Milestone 1 sharing already supports project collaboration)