agentic-stack

Follow me on : @AV1DLIVE for updates/collabs on projects

One brain, many harnesses. A portable .agent/ folder (memory + skills

• protocols) that plugs into Claude Code, Cursor, Windsurf, OpenCode, OpenClient, Hermes, or a DIY Python loop — and keeps its knowledge when you switch.

Based on the article: "The Agentic Stack" · by @AV1DLIVE

---

What this is

Every guide shows the folder structure. This repo gives you the folder structure plus the files that actually go inside — a working portable brain with five seed skills, four memory layers, enforced permissions, a nightly dream cycle, and adapters for seven harnesses.

• Memory — working/, episodic/, semantic/, personal/. Each layer has its own retention policy. Salience-scored retrieval; nightly compression.
• Skills — progressive disclosure. A lightweight manifest always loads; full SKILL.md files only load when triggers match the task. Every skill ships with a self-rewrite hook.
• Protocols — typed tool schemas, a permissions.md that the pre-tool-call hook enforces, and a delegation contract for sub-agents.

Quickstart

# tap + install (one-time — both lines required)
brew tap codejunkie99/agentic-stack https://github.com/codejunkie99/agentic-stack
brew install agentic-stack

# drop the brain into any project — the onboarding wizard runs automatically
cd your-project
agentic-stack claude-code
# or: cursor | windsurf | opencode | openclient | hermes | standalone-python

Already installed? Update to the latest version:

brew update && brew upgrade agentic-stack

Clone instead?

git clone https://github.com/codejunkie99/agentic-stack.git
cd agentic-stack && ./install.sh claude-code

Onboarding wizard

After the adapter is installed, a colorful terminal wizard runs automatically and populates .agent/memory/personal/PREFERENCES.md — the first file your AI reads at the start of every session.

It asks six questions with arrow-key selection, each skippable with Enter:

Question	Default
What should I call you?	(skip)
Primary language(s)?	unspecified
Explanation style?	concise
Test strategy?	test-after
Commit message style?	conventional commits
Code review depth?	critical issues only

Flags:

agentic-stack claude-code --yes          # accept all defaults silently (CI / scripted)
agentic-stack claude-code --reconfigure  # re-run the wizard on an existing project

Edit .agent/memory/personal/PREFERENCES.md any time to add or refine your conventions — the agent re-reads it fresh every session.

Repo layout

.agent/                         # the portable brain (same across all harnesses)
├── AGENTS.md                   # the map
├── harness/                    # dumb conductor + hooks (standalone path)
├── memory/                     # working / episodic / semantic / personal
├── skills/                     # _index.md + _manifest.jsonl + SKILL.md files
├── protocols/                  # permissions + tool schemas + delegation
└── tools/                      # skill loader, memory reflect, budget tracker

adapters/                       # one small shim per harness
├── claude-code/   (CLAUDE.md + settings.json hooks)
├── cursor/        (.cursor/rules/*.mdc)
├── windsurf/      (.windsurfrules)
├── opencode/      (AGENTS.md + opencode.json)
├── openclient/    (system-prompt include)
├── hermes/       (AGENTS.md)
└── standalone-python/  (DIY conductor entrypoint)

docs/                           # architecture, getting-started, per-harness
examples/                       # minimal first_run.py
install.sh                      # one-command adapter install + wizard trigger
onboard.py                      # onboarding wizard entry point
onboard_ui.py                   # ANSI palette, banner, clack-style layout
onboard_widgets.py              # arrow-key prompts (text, select, confirm)
onboard_render.py               # answers → PREFERENCES.md content
onboard_write.py                # atomic file write with backup

Supported harnesses

Harness	Config file it reads	Hook support
Claude Code	CLAUDE.md + .claude/settings.json	yes (PostToolUse, Stop)
Cursor	.cursor/rules/*.mdc	no (manual reflect calls)
Windsurf	.windsurfrules	no (manual reflect calls)
OpenCode	AGENTS.md + opencode.json	partial (permission rules)
OpenClient	system-prompt include	varies by fork
Hermes Agent	AGENTS.md (agentskills.io compatible)	partial (own memory)
Standalone Python	run.py (any LLM)	yes (full control)

Seed skills

• skillforge — creates new skills from recurring patterns
• memory-manager — runs reflection cycles, promotes lessons
• git-proxy — all git ops, with safety constraints
• debug-investigator — reproduce → isolate → hypothesize → verify
• deploy-checklist — the fence between staging and production

How it compounds

1. Skills log every action to episodic memory.
2. memory-manager detects recurring patterns, promotes them to semantic.
3. on_failure flags skills that fail 3+ times in 14 days for rewrite.
4. Self-rewrite hooks on each skill update KNOWLEDGE.md conservatively.
5. git log .agent/memory/ becomes the agent's autobiography.

See docs/architecture.md for the full picture.

Run the dream cycle nightly

crontab -e
0 3 * * * python3 /path/to/project/.agent/memory/auto_dream.py >> /path/to/project/.agent/memory/dream.log 2>&1

auto_dream.py resolves its own paths absolutely, so no cd is needed — cron runs it correctly regardless of working directory.

License

MIT — see LICENSE.

Credits

Design adapted from the author's article on building an agentic stack, plus patterns from Gstack, Claude Code's memory system, and conversations in the agent-engineering community. Built with the hypothesis that harness-agnosticism is the point.