#173173rd of 389 in Skills, plugins and prompts
j-space-cognition-suite
Agent skill with Python gates for long, evidence-backed repo tasks
- Stars
- 3k
- License
- Apache-2.0
- Last commit
- Sep 2026
- Last release
- Sep 2026
- Language
- Python
0 on 21 Jul 20262,956 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
J-Space is a skill directory you copy into a host that loads SKILL.md. It adds thirteen prompt modules and standard-library Python scripts that persist task state, reread source files, detect stale repo maps, and exit nonzero when evidence or reviews are missing. It defines four operating levels (low to xhigh), with child-agent ownership and independent review at xhigh.
Who it is for: Engineers running long repo tasks with skill-capable coding agents
Strengths
- Standard-library Python 3.10+ only; no pip dependencies or background service
- Ship gate exits nonzero when state, source receipts, map or evidence is missing or stale
- Four levels let simple tasks skip persistent setup entirely
- CI covers Windows, Linux and macOS, with unit tests for state and freshness
Weaknesses
- Host must supply tools and agents; installing the files registers no hooks
- Automatic invocation needs manual wiring through host_bridge.py or explicit commands
- README says tests do not show model performance gains
- High and xhigh levels need many manual commands, maps and reports
What it needs
- no GPU
- Needs Python 3.10+, Host that loads SKILL.md
Quick answers
Is j-space-cognition-suite open source?
Yes. Its code is on GitHub under the Apache-2.0 license.
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 |