{"id":44935769,"url":"https://github.com/nvandessel/floop","last_synced_at":"2026-04-01T18:15:36.175Z","repository":{"id":337317871,"uuid":"1142101599","full_name":"nvandessel/floop","owner":"nvandessel","description":"Spreading activation memory for AI coding agents - corrections in, context-aware behaviors out.","archived":false,"fork":false,"pushed_at":"2026-02-26T21:51:18.000Z","size":1738,"stargazers_count":16,"open_issues_count":4,"forks_count":2,"subscribers_count":1,"default_branch":"main","last_synced_at":"2026-02-26T22:06:56.646Z","etag":null,"topics":["ai-agents","behavior-learning","claude-code","cli","golang","mcp-server"],"latest_commit_sha":null,"homepage":"","language":"Go","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/nvandessel.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":"CODE_OF_CONDUCT.md","threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":"SECURITY.md","support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":"AGENTS.md","dco":null,"cla":null}},"created_at":"2026-01-26T00:17:38.000Z","updated_at":"2026-02-26T15:42:24.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/nvandessel/floop","commit_stats":null,"previous_names":["nvandessel/feedback-loop","nvandessel/floop"],"tags_count":36,"template":false,"template_full_name":null,"purl":"pkg:github/nvandessel/floop","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nvandessel%2Ffloop","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nvandessel%2Ffloop/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nvandessel%2Ffloop/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nvandessel%2Ffloop/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/nvandessel","download_url":"https://codeload.github.com/nvandessel/floop/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/nvandessel%2Ffloop/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":29957128,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-02-28T22:53:01.873Z","status":"ssl_error","status_checked_at":"2026-02-28T22:52:50.699Z","response_time":90,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.6:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai-agents","behavior-learning","claude-code","cli","golang","mcp-server"],"created_at":"2026-02-18T07:11:20.244Z","updated_at":"2026-03-08T01:12:02.870Z","avatar_url":"https://github.com/nvandessel.png","language":"Go","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cimg width=\"1600\" height=\"560\" alt=\"img-floop-logo\" src=\"https://github.com/user-attachments/assets/bc695966-8c2b-4956-9b8b-e711333588e4\" /\u003e\n\n# f(eedback)loop\n\n[![CI](https://github.com/nvandessel/floop/actions/workflows/ci.yml/badge.svg)](https://github.com/nvandessel/floop/actions/workflows/ci.yml)\n[![Release](https://img.shields.io/github/v/release/nvandessel/floop)](https://github.com/nvandessel/floop/releases/latest)\n[![Go 1.25+](https://img.shields.io/badge/go-1.25%2B-blue.svg)](https://go.dev/)\n[![License: Apache 2.0](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](LICENSE)\n\nEvery correction you give an AI agent is a lesson that dies at the end of the session. floop makes it stick.\n\nA correction becomes a behavior. Behaviors connect into a graph. The graph uses spreading activation — the same model cognitive science uses to describe how human memory retrieves associations — to find the right behaviors for your current context. Connections strengthen through Hebbian learning. The result is an agent that gets better over time instead of starting from zero every morning.\n\n## Features\n\n- **Learns from corrections** — Tell the agent what it did wrong and what to do instead; floop turns that into a durable behavior\n- **Context-aware activation** — Behaviors fire based on file type, task, and semantic relevance — not a static prompt dump\n- **Spreading activation** — Graph-based memory retrieval inspired by cognitive science (Collins \u0026 Loftus, ACT-R) — triggered behaviors propagate energy to related nodes, pulling in associative context\n- **Vector-accelerated retrieval** — Local embeddings with a tiered vector index (brute-force → HNSW) pre-filter candidates before spreading activation, scaling to thousands of behaviors\n- **Token-optimized** — Budget-aware assembly keeps injected context within limits\n- **Store management** — Stats, deduplication, backup/restore, and graph visualization keep your behavior store healthy\n- **MCP server** — Works with any AI tool that supports the Model Context Protocol\n- **CLI-first** — Every operation available as a command with `--json` output for agent consumption\n\n## Quick Start\n\n### Install\n\n```bash\n# Homebrew (macOS/Linux)\nbrew install nvandessel/tap/floop\n\n# Go\ngo install github.com/nvandessel/floop/cmd/floop@latest\n```\n\n### Teach your agent something\n\n```bash\ncd your-project \u0026\u0026 floop init\n\n# Capture a correction\nfloop learn --right \"Always use structured logging, never fmt.Println\"\n\n# See what floop learned\nfloop list\n```\n\n### See it activate\n\n```bash\n# Check what behaviors fire for your current context\nfloop active --file src/main.go --task development\n```\n\nFor a hands-on walkthrough, see the [5-minute tutorial](docs/WALKTHROUGH.md).\n\n## Integrate with your AI tool\n\nAdd floop as an MCP server so your AI tool loads behaviors automatically.\n\n**Claude Code** (`~/.claude/settings.json`):\n```json\n{\n  \"mcpServers\": {\n    \"floop\": {\n      \"command\": \"floop\",\n      \"args\": [\"mcp-server\"]\n    }\n  }\n}\n```\n\nSee [docs/integrations/](docs/integrations/) for setup guides for Cursor, Windsurf, Copilot, and more.\n\n### Store management\n\n```bash\nfloop stats                          # Check behavior store health\nfloop deduplicate --dry-run          # Find duplicate behaviors\nfloop connect \u003csrc\u003e \u003ctgt\u003e --kind similar-to  # Link related behaviors\n```\n\n## How It Works\n\n```\n You correct          floop extracts         Behaviors stored         Spreading activation        Context injected\n your agent     →     a behavior       →     in a graph         →    finds relevant nodes   →    into next session\n      ↑                                                                                               │\n      └───────────────────────── agent improves, cycle repeats ────────────────────────────────────────┘\n```\n\nWhen you correct your AI agent, floop captures the correction and extracts a **behavior** — a reusable rule with context conditions. Behaviors are stored as nodes in a graph, connected by typed edges (similar-to, learned-from, requires, conflicts).\n\nWhen you start a session, floop builds a context snapshot from your current file, task, and project. It uses **spreading activation** to propagate energy through the graph from matching nodes. Energy cascades outward through associations, pulling in related behaviors — like the brain activating related memories through associative networks. The result is a focused set of behaviors tuned to your current work.\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"docs/images/graph-view.png\" alt=\"floop behavior graph — 55 nodes, 282 edges\" width=\"720\"\u003e\n  \u003cbr\u003e\n  \u003cem\u003eInteractive behavior graph built from real corrections — nodes are behaviors (colored by type), edges are relationships.\u003c/em\u003e\n\u003c/p\u003e\n\n## Documentation\n\n**Get started:**\n- [5-minute walkthrough](docs/WALKTHROUGH.md) — Hands-on toy project\n- [Integration guides](docs/integrations/) — Claude Code, Cursor, Windsurf, and more\n\n**Reference:**\n- [CLI reference](docs/CLI_REFERENCE.md) — All commands and flags\n- [Usage guide](docs/FLOOP_USAGE.md) — MCP and CLI workflows\n\n**Deep dives:**\n- [Similarity pipeline](docs/SIMILARITY.md) — Deduplication and matching\n- [Local embeddings](docs/EMBEDDINGS.md) — Semantic retrieval\n- [Research \u0026 theory](docs/SCIENCE.md) — Cognitive science background\n- [Origin story](docs/LORE.md) — How floop came to be\n- [Contributing](CONTRIBUTING.md)\n\n## Project Status\n\nfloop is a working tool I use daily to build floop itself (160+ learned behaviors and counting). It's a hobby project built in my free time — actively maintained, tested (90%+ coverage on core packages, race-clean), and used in production on my own workflows. The CLI and MCP interfaces are stable; internals may evolve between minor versions. Contributions and feedback are welcome.\n\n## License\n\n[Apache License 2.0](LICENSE)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnvandessel%2Ffloop","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fnvandessel%2Ffloop","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fnvandessel%2Ffloop/lists"}