Skip to main content
Glama

Athena

A project-aware, capability-driven advisory framework for AI coding assistants.

Athena is an AI Consultation Runtime. Cursor Composer owns execution; Athena owns consultation.

Features

  • Single MCP tool: consult(advisor_type, question, session_context, mode)

  • Capability-driven routing across Anthropic, OpenAI, and Ollama providers

  • Project memory with SQLite storage and git-based incremental indexing

  • Structured advisory responses for Composer to apply

Related MCP server: repowise

Quickstart

pip install -e ".[dev,providers]"
export ANTHROPIC_API_KEY=your-key
python -m athena.mcp_server.server

Add the plugin MCP config from plugin/.mcp.json to your project's .cursor/mcp.json.

Documentation

Engineering specifications live in docs/specs/README.md.

Development

pip install -e ".[dev,providers]"
python3 -m pytest -q
python3 -m ruff check athena tests benchmarks
python3 benchmarks/run_baseline.py

See docs/extension-guide.md for adding providers, advisor profiles, capabilities, and strategy modes.

License

MIT — see LICENSE.

Available Tools

5 tools
athena_indexC

Trigger project memory indexing

Inputs: mode, paths.

athena_rememberC

Persist a design decision to project memory

Inputs: decision (required), tags, affects, rationale.

athena_statusA

Return Athena project memory and provider status

Inputs: no named parameters.

consultC

Consult Athena for expert engineering advice. Advisor types include architecture, security, debugging, planning, review, optimization, documentation.

Inputs: question (required), advisor_type (required), mode, constraints, +1 more.

get_consultation_historyC

Retrieve prior consultations

Inputs: limit, since, advisor_type, consultation_id.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    B3/5.0

    Scored across 5 tools

    Disambiguation5/5

    Each tool has a distinct purpose: indexing, remembering decisions, checking status, consulting, and viewing history. No overlap in functionality.

    Naming Consistency3/5

    Three tools follow the 'athena_verb' pattern, but 'consult' and 'get_consultation_history' break the pattern with different prefixes, introducing inconsistency.

    Tool Count5/5

    Five tools is a well-scoped set for a project memory and expert advisor server, covering core operations without being excessive.

    Completeness4/5

    The surface covers essential memory and consulting operations, but could include update/delete for memories or clearing consultation history for full lifecycle coverage.

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      B
      quality
      D
      maintenance
      An MCP server that enables AI-powered IDEs to implement a structured development workflow from requirements gathering to code implementation, guiding users through goal collection, requirements specification, design documentation, task planning, and execution.
      9
      6 npm
      22
      MIT
    • A
      license
      A
      quality
      B
      maintenance
      Provides AI coding agents with five intelligence layers (dependency graph, git history, documentation, architectural decisions, code health) via nine MCP tools, enabling deep codebase understanding and reducing exploration cost.
      5
      6,899 PyPI
      8,180
      AGPL 3.0
    • A
      license
      Not graded
      quality
      C
      maintenance
      An MCP server that extends AI coding assistants with deterministic, algorithmic capabilities such as code analysis, fault localization, and formal verification, enabling an autonomous engineering team within the IDE.
      MIT