⣠⣾⣷⣶⣦⣤⣤⣄⣠⣄⣀ ⢀⣀⣀ ⢀⣴⣿⣿⠿⣋⣭⣭⣯⣭⣍⣭⣿⣟⠛⠛⠿⠿⣿⣷⣄ ⢀⣴⣾⡟⢻⣿⡟⠁⣼⣿⠏⣵⢻⣿⣻⣿⣿⢿⡻⣿⣿⣶⡌⢿⣿⣷⣦⣤⡄ ⣤⣶⣾⣿⣿⠏ ⠈⢿⣄ ⢹⣏⠠⠟⣾⣿⣿⣿⣿⣿⠷⣏⣼⠟⢡⣿⡟⠋⢻⣿⣿⡄ ⠈⣿⣿⣿⣿⡆ ⣽⢧⡘⠈⠳⣦⣍⠛⠛⢦⣉⣴⣛⣫⣭⣴⡟⠋ ⣾⣿⣿⡿ ⢀⠹⣿⣿⣿⣷⣤⡄ ⠋ ⠙⢆ ⣠⠴⠟⠛⣛⣛⣛⠟⠋⠁⠺⡇ ⣀⣴⣿⣿⡟⠁ ⠈⣀⠈⠛⠷⠿⣿⣿⣷⣤⣀ ⢠⠋ ⠈⠉⠉ ⣠⣴⣥⠾⠛⠉⣰⣿⣷ ⠹⣯⣝⠛⠛⠷⢶⣤⣤⣀ ⢀⡠⠖⠋⠉⢉⣀⣀⣴⣾⣿⠿⠟⠃ ⠠⠦ ⠁ ⠖ ⠘⠻⢿⣦⣄⡀ ⠉⠛⢦⠠⢊⠤⠴⢒⣛⣛⣩⣽⡿⠟⠁⢀⡀ ⠲⠶⣦⠴⠶⠶⠶⠶⡶⠶⢶⣤⣄⡀⠨⠭⠽⠟⣓⢦⣀⠈⢇⡥⠖⠛⠋⠉⠉⠉ ⠈ ⢠⡤ ⠈⢷ ⠐⠂⢤⣽⣄ ⠰⡎⠙⠳⣄⡀ ⠈⢣⠘⢦⠋⣀⡬⠟⠛⠛⠉⢀⣀⣀⣠⡤⠄⠃ ⠈⢳⣀⡒⠉⠉⣉⠙⡲⣽⣄ ⣏⠳⡄ ⠘⡇ ⡾⠁ ⢀⡤⠖⣻⣿⡏⢡⡎ ⠰⠄ ⠛⠻⢦⣄⣉⡁⣀⣀⣈⣙⣺⣌⡇⢠⢀⡇⡾ ⣴⣿⡷⠊ ⢲⣠⠟ ⠈⠉ ⠈⠳⡄⣸⢱⠇⢀⣰⣯⣭⣥⠭⠾⠛⠃ ⡷⠡⡯⢖⠉ ⢠⠤ ⡠⢊⡴⠤⠂⠃ ⠒ ⢀⡴⢪⠔⣉⠔⠋ ⠐⠈
Gentle-AI 2.5.1 — Ecosystem, Frameworks, Workflows for AI coding agents
You installed Claude Code, OpenCode, Cursor or Codex. It writes code, but it forgets everything between sessions, has no opinion about how your project works, and gives you no way to check its work beyond reading every line yourself.
Gentle-AI equips the agent runtime already on your machine with persistent memory, a planning workflow, a curated skill library, MCP tool servers, model routing, a teaching-oriented persona and an optional evidence-based review step. Pick a topic below to learn more.
Menu
j/k: navigate • enter: select
Gentle-AI is an ecosystem configurator. It takes the agent runtime already on your machine and equips it with persistent memory, a planning workflow, a curated skill library, MCP tool servers, model routing, a teaching-oriented persona, and an optional evidence-based review step.
Before: "I installed an AI agent, but it's just a chatbot that writes code." After: the agent remembers your decisions, follows your project's conventions, picks a working style that matches the size of the task, and can produce reviewable evidence for what it changed.
It never installs an AI agent for you. It configures runtimes that are already present; if it cannot detect an agent you selected, it refuses and prints the exact command you would run yourself.
enter: select • esc: back
Gentle-AI installs a set of components. You pick them individually, or take a preset.
enter: select • esc: back
Gentle-AI configures each agent using that agent's own native features, so capabilities differ. The delegation model tells you whether the agent can hand work to focused sub-agents or runs everything in one conversation.
enter: select • esc: back
Prerequisites: Node.js 18+ and npm, Git 2.38+, Go 1.25.10+ on Windows, and your AI agent already on your PATH. Gentle-AI configures it; it does not install it.
Then install the binary, launch the TUI to select your agents, components (or a preset) and persona, and verify with a read-only health report. Your previous configs are snapshotted first.
$ brew install gentleman-programming/tap/gentle-ai $ gentle-ai $ gentle-ai doctor
enter: select • esc: back
This is the core idea, and it applies to every configured agent, even if you never enable SDD or review. You ask for an outcome; the agent picks the smallest route that gets there. It does not escalate ceremony because a task feels big.
enter: select • esc: back
SDD is a planning workflow for substantial features. Instead of jumping straight into code, the agent explores the codebase, proposes an approach you approve, writes requirements, designs the architecture, breaks it into ordered tasks, implements, then independently verifies the result against what was agreed.
Use it when the work is ambiguous enough that written artifacts would genuinely reduce that ambiguity. For a bug fix or a small feature, it is overhead.
You do not learn the phases. Say "use SDD" and the agent starts the workflow, or accept it when the agent offers. You review and approve at the decision points.
$ /gentle-sdd-new $ /gentle-sdd-continue
enter: select • esc: back
An agent tells you "I fixed it and the tests pass." You have no way to check that claim except reading the whole diff yourself. RDD replaces agent narration with evidence the system can derive independently.
Once implementation finishes, RDD freezes the candidate and picks review effort from evidence, not from size: low risk gets a structural readback, medium risk one focus lens, high risk all four lenses (Risk, Resilience, Readability, Reliability) with a cost forecast. At most one bounded correction, then a read-only validator checks it.
RDD is opt-in and off by default. Nothing happens until you enable it.
$ gentle-ai review mode enable --scope global $ gentle-ai review mode status
enter: select • esc: back
Refresh the binary and its managed agent assets together. Sync is not optional after an upgrade: it refreshes the assets Gentle-AI writes into your agents.
Every install, sync and upgrade snapshots your config files. Backups are compressed, deduplicated and auto-pruned to the five most recent; pin important ones from the TUI.
$ gentle-ai upgrade $ gentle-ai sync
enter: select • esc: back
Open the helper agent and ask anything about Gentle-AI or the rest of the ecosystem.
enter: select • esc: back
Warning: this space simulates gentle-pi for the landing page; it is not a real install.
Skill registry refreshed (52 skills)
pi update --extensions🐘 SYSTEM ONLINEMEM: OK 100%
███████ ███ ██ ██████ ██████ █████ ███ ███ ██ ████ ██ ██ ██ ██ ██ ██ ████ ████ █████ ██ ██ ██ ██ ███ ██████ ███████ ██ ████ ██ ██ ██ ██ ██ ██ ██ ██ ██ ██ ██ ██ ██ ██ ███████ ██ ████ ██████ ██ ██ ██ ██ ██ ██
> engram 1.20.0 — An elephant never forgets
Projects
...and 70 more projects
Actions
j/k navigate • enter select
engram, neuroscience: the physical trace of a memory in the brain. Your AI coding agent forgets everything when the session ends. Engram gives it a brain.
A Go binary with SQLite + FTS5 full-text search, exposed through CLI, HTTP API, MCP and an interactive TUI. It works with any MCP-compatible agent, including Claude Code, OpenCode, Gemini CLI, Codex, VS Code (Copilot), Antigravity, Cursor and Windsurf.
No Node.js, Python or Docker required: one binary, one SQLite file at ~/.engram/engram.db.
enter select • esc back
Treat Engram as a curated project memory, not a transcript sink. The operating contract an agent follows through a session:
enter select • esc back
Start with the intent, then reach for the tool. Availability can vary by MCP profile.
enter select • esc back
Install the binary, then run the setup command for the agent you use and restart it. engram setup writes the MCP and integration configuration; the usual stdio-only setup needs no running server.
Claude Code installs it as a plugin; Pi, OpenCode, Gemini CLI, Codex, Antigravity CLI, Windsurf, Qwen Code, Kiro, Cursor, VS Code Copilot and Kilo Code use engram setup. Any other MCP-compatible agent can be configured manually.
$ brew install gentleman-programming/tap/engram $ claude plugin marketplace add Gentleman-Programming/engram && claude plugin install engram $ engram setup pi $ engram setup opencode
enter select • esc back
Engram keeps memory local by default. The local SQLite database is authoritative.
Git Sync exports portable compressed chunks for sharing memory across machines. Engram Cloud is optional, project-scoped replication and shared access with browser visibility.
Project-aware reads use the canonical current project when no selector is supplied: an explicit project, then ENGRAM_PROJECT, then cwd detection.
$ engram cloud upgrade doctor --project <project> $ engram cloud upgrade bootstrap --project <project>
enter select • esc back
The dashboard you are looking at. Navigate with j/k, use enter to drill in, c to copy content to the clipboard, / to search and esc to go back.
Search memories, browse recent observations and sessions, set up an agent plugin and manage cloud sync settings without leaving the terminal.
$ engram tui
enter select • esc back
Open the helper agent and ask anything about Engram or the rest of the ecosystem.
enter select • esc back
Gentleman Guardian Angel code review
Install: brew install gentleman-programming/tap/gga
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Gentleman Guardian Angel v2.10.1
Provider-agnostic code review using AI
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
USAGE:
gga <command> [options]
COMMANDS:
run [--no-cache] Run code review on staged files
install Install git pre-commit hook (default)
install --commit-msg
Install git commit-msg hook (for commit message validation)
uninstall Remove git hooks from current repo
config Show current configuration
init Create a sample .gga config file
cache clear Clear cache for current project
cache clear-all Clear all cached data
cache status Show cache status
help Show this help message
version Show version
RUN OPTIONS:
--no-cache Force review all files, ignoring cache
--ci CI mode: review files changed in last commit (HEAD~1..HEAD)
--pr-mode PR mode: review all files changed in the full PR
--diff-only With --pr-mode: send only diffs (faster, cheaper)
CONFIGURATION:
Create a .gga file in your project root or
~/.config/gga/config for global settings.
CONFIG OPTIONS:
PROVIDER AI provider to use (required)
Values: claude, gemini, codex, opencode, cursor[:model], kilo[:model], kiro, ollama:<model>
FILE_PATTERNS File patterns to review (default: *)
EXCLUDE_PATTERNS Patterns to exclude from review
RULES_FILE File containing review rules (default: AGENTS.md)
STRICT_MODE Fail on ambiguous AI response (default: true)
TIMEOUT Max seconds to wait for AI response (default: 300)
EXAMPLES:
gga init # Create sample config
gga install # Install pre-commit hook (default)
gga run # Run review (with cache)
gga run --pr-mode # Review full PR changes
gga cache status # Show cache info
Type help or click a suggestion above.
The complete development environment installer
Install: brew install Gentleman-Programming/tap/gentleman-dots
What the installer sets up
Vim Mastery Trainer (RPG mode, launched from the installer menu)
7 modules with 15 progressive lessons each, practice mode with intelligent exercise selection, boss fights and XP tracking.
| Module | Keys covered |
|---|---|
| Horizontal Movement | w e b f t 0 $ ^ |
| Vertical Movement | j k G gg { } |
| Text Objects | iw aw i" a( it at |
| Change & Repeat | d c dd cc D C x |
| Substitution | r R s S ~ gu gU J |
| Macros & Registers | qa @a @@ "ay "+p |
| Regex / Search | / ? n N * # |
gentleman-cute palette
~/.config/starship.toml (excerpt)
[palettes.gentleman-cute] pink = "#FFB1DD" mauve = "#D7A0B8" blue = "#F095C8" [git_branch] format = "-> [$symbol$branch]($style) " symbol = " " [character] success_symbol = "[ ](fg:green)" error_symbol = "[ ](fg:red)"
~/.config/fish/conf.d/gentleman-cute.fish (excerpt)
set -g fish_color_command --bold $powder_blue set -g fish_color_keyword --bold $accent set -g fish_color_quote $mint set -g fish_color_error --bold $error set -g fish_color_autosuggestion --dim $muted
~/.config/herdr/profiles/gentleman-cute.toml (excerpt)
[theme.custom] panel_bg = "#000000" accent = "#F095C8" mauve = "#FFB1DD" green = "#B4E7C7" [ui] accent = "#F095C8"
Warning: this space simulates gentle-pi for the landing page; it is not a real install.
Skill registry refreshed (52 skills)
pi update --extensions