#198198th of 389 in Skills, plugins and prompts
powercontext
Server-side memory and handoff store for AI coding agents across sessions
- Stars
- 1.6k
- License
- Apache-2.0
- Last commit
- Nov 2026
- Last release
- Nov 2026
- Language
- Python
620 on 21 Apr 20261,596 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
PowerContext is a context service that stores decisions, constraints and task state as revisioned artifacts (Memory, Handoff, Experience, Skill) outside any single chat. It exposes HTTP, MCP, a Python client, a CLI and a Dashboard, with plugins for Codex and Claude Code. Experience and Skill candidates go through a review inbox before reuse, and recall is capped by a byte budget.
Who it is for: Teams handing work between coding agents and people with shared context
Strengths
- Artifacts addressed as FAMILY/ID@REVISION; exact historical revisions stay readable
- Basic mode saves and recalls memory with no server-side model API
- Same Scope shared across Codex, Claude Code and other hosts, and people
- Review inbox with expected_version checks before Experience or Skill is reused
Weaknesses
- Automatic extraction and Topic Memory need separate Generation and Embedding API credentials
- Windows support is experimental; macOS and Linux only
- LoCoMo Plus scores were supplied by the owner; run artifacts are not public
- Desktop app is a preview for internal validation, not signed for distribution
What it needs
- no GPU
- Docker + Compose
- Needs Python 3.11+, Git, uv, SQLite or embedded seekdb or OceanBase
- Models: Codex, Claude Code, MiniMax Code, DeepSeek Harness, GPT-4o
Quick answers
Is powercontext open source?
Yes. Its code is on GitHub under the Apache-2.0 license.
Can you run powercontext with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Also in Skills, plugins and prompts
See all 389| Rank | Project | Score |
|---|---|---|
| 1 | claude-memPersistent cross-session memory plugin for Claude Code and other coding agents | 97 out of 100 |
| 2 | impeccableDesign skill with 24 slash commands for AI coding agents' frontend output | 97 out of 100 |
| 3 | gstackMarkdown slash-command skills that give Claude Code planning, review, QA and ship roles | 96 out of 100 |
| #4 | rufloAgent harness adding swarms, memory and hooks to Claude Code and Codex | 96 out of 100 |
| #5 | cavemanShortens coding-agent replies and compresses tool output before the model reads it | 95 out of 100 |