Skip to main content

Project Memory: repo-scoped memory engine for AI agents

Project description

Project Memory — Repo-Scoped Memory Engine

This project is a repo-scoped memory engine, including:

  • CLI (project-memory) via Typer
  • SQLite memory database under .project-memory/
  • File ingestion plus typed memory for notes, learnings, tasks, and plans
  • Immutable history snapshots for notes, learnings, tasks, and plans
  • Keyword search across indexed files and typed memory
  • MCP HTTP server over streamable HTTP
  • GitHub Actions CI/release/publish workflows

Quickstart

python -m pip install -e .
project-memory init
project-memory index
project-memory search "your query"
project-memory remember deploy "Run migrations before restart"
project-memory plan create branching "Use feature branches" --type protocol
project-memory serve-mcp

Current behavior

  • project-memory index indexes repository text files only. Reindexing cleans up stale file documents, but it does not delete notes, learnings, tasks, or plans.
  • project-memory search is keyword-only today. Search results can include both indexed files and typed memory.
  • project-memory export / project-memory import round-trip task and plan status, including archived plans.
  • Typed memory writes create immutable history snapshots, and entries can be diffed or restored via project-memory history ....
  • Embedding configuration commands exist, but embedding-backed indexing and query-time hybrid search are not implemented yet.

Commands

  • project-memory init - initialize .project-memory/project_memory.db
  • project-memory index - ingest repository text files into SQLite
  • project-memory search QUERY - perform a keyword search across files and typed memory
  • project-memory remember KEY CONTENT - store a note
  • project-memory learn KEY CONTENT - store a learning
  • project-memory task ... - manage tracked tasks
  • project-memory plan ... - manage implementation plans and protocols
  • project-memory export / project-memory import - round-trip memory through MEMORY.md
  • project-memory history ... - list, inspect, diff, and restore typed memory revisions
  • project-memory serve-mcp - run the MCP server at http://127.0.0.1:8000/mcp/

Project structure

  • src/project_memory/ core implementation
  • .project-memory/ runtime storage
  • .github/workflows CI/CD

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

project_memory-0.2.2.tar.gz (132.4 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

project_memory-0.2.2-py3-none-any.whl (29.1 kB view details)

Uploaded Python 3

File details

Details for the file project_memory-0.2.2.tar.gz.

File metadata

  • Download URL: project_memory-0.2.2.tar.gz
  • Upload date:
  • Size: 132.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for project_memory-0.2.2.tar.gz
Algorithm Hash digest
SHA256 776b13d100eaee8c2b18702c926840c5024d1995aa0e89b7d656bf7353296975
MD5 1a0632424486e38ba6c9c6bd33585d4c
BLAKE2b-256 1af4fcd1600b094ead06dc380c82d89866203ccf7f089d772cdf8792f7fc3ae1

See more details on using hashes here.

Provenance

The following attestation bundles were made for project_memory-0.2.2.tar.gz:

Publisher: release.yml on jflamb/project-memory

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file project_memory-0.2.2-py3-none-any.whl.

File metadata

  • Download URL: project_memory-0.2.2-py3-none-any.whl
  • Upload date:
  • Size: 29.1 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for project_memory-0.2.2-py3-none-any.whl
Algorithm Hash digest
SHA256 cc3f99863850b9720a9dde77ae9fdbe992d86c75f1ba746c9a2a55447c916b57
MD5 50a722cf3744349cf652a099e5793fd4
BLAKE2b-256 fa34104a0283eb77254505a9b8888e49213b557960792ee4c764726e1f17a1bc

See more details on using hashes here.

Provenance

The following attestation bundles were made for project_memory-0.2.2-py3-none-any.whl:

Publisher: release.yml on jflamb/project-memory

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page