SQLite Project Memory MCP

SQLite-backed MCP server for storing project memory as a graph-friendly relational core.

Why this project

• Keeps authoritative state in SQLite (single source of truth).

• Provides graph semantics (entities, relationships, content, tags).

• Supports structured reads (query_view) and explicit generated markdown views.

• Designed for AI-agent-safe workflows.

Core model

• entities

• attributes

• relationships

• content

• tags

• events, snapshots, snapshot_entities (audit/history)

Supported sync-document targets

• Core documents: architecture, decisions, plan, notes, roadmap

• Extended docks: kpi, okr, strategy, risk, issue, epic, story, feature, milestone, release, dependency, objective, initiative, metric, capability, assumption, problem_statement, retrospective, action_item

Install

Clone the repo into a sqlite-mcp subfolder of your project, then run the installer once. The script detects its location, places all runtime files inside Project Memory/, and writes .vscode/mcp.json.

No Python installation required. The installer downloads uv and uses it to fetch a pinned CPython 3.12.9 interpreter automatically. The interpreter is cached in Project Memory/.uv/python/ and the virtual environment is created in Project Memory/.venv. If the uv download fails (e.g. behind a firewall) the installer falls back to any Python 3.11+ already on the machine.

Important: open VS Code on your project root (the parent folder), not on the sqlite-mcp subfolder. The MCP config is written to the project root at .vscode/mcp.json.

Windows (PowerShell)

# from your project root
git clone https://github.com/WebRTCGame/SQLITE-MCP.git sqlite-mcp
.\sqlite-mcp\install.ps1

Optional flag:

• -LogFile install.log — save a full transcript for debugging

• -AppendInstructions — append snippet to the suggested instructions file (idempotent)

Linux / macOS

# from your project root
git clone https://github.com/WebRTCGame/SQLITE-MCP.git sqlite-mcp
chmod +x ./sqlite-mcp/install.sh
./sqlite-mcp/install.sh

Optional flag:

• --log-file install.log — save a full transcript for debugging

• --append-instructions — append snippet to the suggested instructions file (idempotent)

To update: after a successful nested install, the scripts live under Project Memory/ because the sqlite-mcp checkout is moved there. Re-run the installer from your project root with Project Memory\install.ps1 (Windows) or Project Memory/install.sh (Linux/macOS). For in-place/developer installs, re-run the same command.

pip only (developer / advanced)

Used when you want to manage the environment yourself (requires Python 3.11+):

python -m venv .venv
source .venv/bin/activate   # or .venv\Scripts\Activate.ps1 on Windows
pip install -e .

Or with uv directly (no system Python needed — uv downloads Python automatically):

uv venv --python 3.12.9 .venv
uv pip install -e .

Full example:

./sqlite-mcp/install.sh

Start server

& ".\Project Memory\.venv\Scripts\python.exe" -m sqlite_mcp_server

Paths

• Project Memory/.venv — virtual environment (pinned Python 3.12.9)

• Project Memory/.uv/bin/uv[.exe] — uv runtime manager (downloaded by installer)

• Project Memory/.uv/python/ — pinned CPython interpreter (fetched by uv on first run)

• Project Memory/pm_data/project_memory.db

• Project Memory/pm_exports

• .vscode/mcp.json

• .vscode/settings.json may also be created automatically by Copilot/VS Code with chat.mcp.serverSampling entries for sqlite-project-memory; this is expected.

To override the pinned Python version:

$env:SQLITE_MCP_PYTHON_VERSION = "3.13.0"; .\sqlite-mcp\install.ps1

SQLITE_MCP_PYTHON_VERSION=3.13.0 ./sqlite-mcp/install.sh

CLI tools

sqlite-project-memory-admin bootstrap-self --repo-root .
sqlite-project-memory-admin project-state
sqlite-project-memory-admin health
sqlite-project-memory-admin sync-document architecture --input-path architecture.md
sqlite-project-memory-admin sync-document decisions --input-path decisions.md
sqlite-project-memory-admin sync-document roadmap --input-path roadmap.md
sqlite-project-memory-admin export-views --user-requested --request-reason "User asked for a roadmap export" --require-existing-dir exports todo roadmap architecture
sqlite-project-memory-admin export-views --user-requested --request-reason "User asked for refreshed generated docs" --force todo roadmap architecture
sqlite-project-memory-admin export-json --output-path exports/project_memory.snapshot.json
sqlite-project-memory-admin import-json --input-path exports/project_memory.snapshot.json

Configuration

Environment variables:

• SQLITE_MCP_PROJECT_ROOT

• SQLITE_MCP_DB_PATH

• SQLITE_MCP_EXPORT_DIR

• SQLITE_MCP_TRANSPORT (stdio or streamable-http)

• SQLITE_MCP_LOG_LEVEL (INFO default)

• SQLITE_MCP_LOG_FORMAT (json or text)

Post-install hook

• Create .scripts/post_install.ps1 (PowerShell) or .scripts/post_install.sh.

• The install script executes the hook if present.

Quick start (Windows)

1. Clone the repo into your project.

2. Run the installer:

   git clone https://github.com/WebRTCGame/SQLITE-MCP.git sqlite-mcp
   .\sqlite-mcp\install.ps1

   After install, ensure VS Code is opened at the parent project root (for example C:\CODE\TestProject), not C:\CODE\TestProject\sqlite-mcp.

3. Activate runtime venv:

   & ".\Project Memory\.venv\Scripts\Activate.ps1"

4. Run self-check:

   sqlite-project-memory-admin --db-path "Project Memory/pm_data/project_memory.db" project-state
   sqlite-project-memory-admin --db-path "Project Memory/pm_data/project_memory.db" health

5. Start the server:

   python -m sqlite_mcp_server

Quick start (Linux/macOS)

1. Clone the repo into your project.

2. Run the installer:

   git clone https://github.com/WebRTCGame/SQLITE-MCP.git sqlite-mcp
   chmod +x ./sqlite-mcp/install.sh
   ./sqlite-mcp/install.sh

   After install, ensure VS Code is opened at the parent project root, not the sqlite-mcp subfolder.

3. Activate runtime venv:

   source "Project Memory/.venv/bin/activate"

4. Run self-check:

   sqlite-project-memory-admin --db-path "Project Memory/pm_data/project_memory.db" project-state
   sqlite-project-memory-admin --db-path "Project Memory/pm_data/project_memory.db" health

5. Start the server:

   python -m sqlite_mcp_server

Notes

• render_markdown_views/export_markdown_views require user_requested=true and request_reason.

• sync-document is anchored (no unstructured free-text record in roadmap state).

• project_state and health are primary diagnostics.

AI-first design

• Write entities/content through MCP tools.

• Generate docs only on explicit user request.

Add AI instructions to your project

After running the installer, paste the snippet below into your AI instructions file (copilot-instructions.md, AGENTS.md, CLAUDE.md, or equivalent). The installer also prints this snippet to the terminal as a reminder.

## SQLite Project Memory

This project uses the **`sqlite-project-memory` MCP server** as the single source of truth for all project state.

### Mandatory rules

- **NEVER** create, modify, or delete `.md` files for notes, tasks, decisions, roadmaps, or architecture.
- **ALWAYS** use `sqlite-project-memory` MCP tools to store and retrieve project knowledge.
- `README.md` is the **only** markdown file you may edit directly.
- Files in `Project Memory/pm_exports/` are read-only derived artifacts — never edit them directly.

### Tool usage guide

| Intent | Tool to use |
|--------|-------------|
| Add a task / todo | `create_entity` (entity_type: `task`) |
| Record a decision | `create_entity` (entity_type: `decision`) |
| Document a component | `create_entity` (entity_type: `component`) + `write_content` |
| Add a note to an item | `write_content` or `append_content` |
| Query project state | `query_view`, `list_entities`, `get_entity` |
| Search knowledge | `search_content` |
| Link two items | `add_relationship` or `connect_entities` |
| Export to markdown | `export_markdown_views` with `user_requested: true` — only when explicitly asked |

### First action each session

Before making changes, call `get_project_context` to confirm the database path and project root, then call `get_recent_activity` or `query_view` to orient yourself.

Usage gates checklist

For best reliability, confirm all of the following:

1. .vscode/mcp.json contains the sqlite-project-memory server.

2. VS Code MCP approval/trust prompt was accepted for this workspace.

3. Chat is in Agent mode (not regular chat/edit mode).

4. Project Memory agent is available in the agents dropdown.

5. /sqlite-project-memory skill is available in slash commands.

6. Your project instructions file includes the SQLite Project Memory snippet.

7. Start each session with get_project_context then get_recent_activity or query_view.

The installer prints a Usage Gates Report with PASS or ACTION REQUIRED for the gates it can validate automatically. If tools still do not appear, first fully reload or restart VS Code and start a new Agent chat session. If the server is not already running after restart, run MCP: Start Server from the Command Palette, select sqlite-project-memory, and restart the chat session.