Back: Catalog

gentle-ai

Gentle-AI is a deterministic engineering environment that provides permanent memory, workflow, and evidence mechanisms for existing AI coding assistants. It ensures that the project context with Engram is maintained between sessions and manages workloads with Organic-Driven Development (ODD). Claude Code integrates with many popular AI tools such as Cursor, Pi, and Codex.

GE

gentle-ai

75

Gentleman-Programming · AI Coding Assistants

STARS8.2K
FORKS874
COMMITS0
CONTRIBUTORS0
VERSIONv3.7.0
UPDATEDtoday
GoAI Coding AgentWorkflowDeterministicMIT

Common commercial products in this category: GitHub Copilot, Cursor, Tabnine

What can it do?
  • With the engram component, you prevent the AI assistant from losing intersessional context.
  • With the Organic-Driven Development (ODD) method, you keep small jobs light and document big jobs.
  • You freeze and inspect the review evidence of work completed with Receipt-Driven Development (RDD).
User reviews
  • ↓ NegativeGitHub Issue

    bug(review 2.8.2): correction_required capture's own status_continuation classifies the just-minted authority corrupted_

    “Automated provider-defect report. Observed evidence only; no root cause asserted. ## Environment - Build: `gentle-ai 2.8.2` (no `-rc.` / `-main.` tag → stable evidence channel) - OS / arch: darwin / arm64 - Client: claude-code (in-process reviewer capture, `--agent`, no…”

  • ↓ NegativeGitHub Issue

    review status fails before authority evaluation after bounded correction

    “## Problem After a bounded correction in an active negotiated review, the read-only status operation fails safely before native authority evaluation. Repeating the status operation produces the same failure, so the consumer cannot obtain the provider-owned next transition and…”

Website GitHub

My note

Alternative pages

GitHub Copilot alternativesCursor alternativesTabnine alternatives

More in: AI Coding Assistants