# loom

> A fully autonomous (no human-in-loop) agentic based project design and coding machine

- **URL**: https://www.freshcrate.ai/projects/loom
- **Author**: jordanhubbard
- **Category**: Uncategorized
- **Latest version**: `v0.1.26` (2026-03-03)
- **License**: MIT
- **Source**: https://github.com/jordanhubbard/loom
- **Language**: Go
- **GitHub**: 98 stars, 13 forks
- **Registry**: github
- **Tags**: `go`

## Description

A fully autonomous (no human-in-loop) agentic based project design and coding machine

## Recent releases

| Version | Date | Urgency | Changes |
| --- | --- | --- | --- |
| `v0.1.26` | 2026-03-03 | Low | ## 🎉 Loom v0.1.26  ### 📊 Statistics - **Commits since v0.1.25**: 3 - **Test Status**: ✅ All tests passed - **Linters**: ✅ Passed (Go, JavaScript, YAML, Docs, API validation)  ### 📝 Changes  - chore: broaden gitignore to cover all .beads/ runtime dirs under internal/ and app/ - fix: update stale test assertions after dispatcher deprecation and repl error change - docs: Update CHANGELOG for v0.1.25 release  ### 🔗 Links - [Full Changelog](https://github.com/jordanhubbard/Loom/compare/v0.1.25... |
| `v0.1.25` | 2026-03-03 | Low | ## 🎉 Loom v0.1.25  ### 📊 Statistics - **Commits since v0.1.24**: 3 - **Test Status**: ⚠️ Some tests failed (see test output) - **Linters**: ✅ Passed (Go, JavaScript, YAML, Docs, API validation)  ### 📝 Changes  - chore: gitignore internal/beads/.beads/ runtime data (modified by every test run) - fix: validate log handler params before nil-checking logManager; guard nil Dispatcher - docs: Update CHANGELOG for v0.1.24 release  ### 🔗 Links - [Full Changelog](https://github.com/jordanhubbard/Loom |
| `v0.1.24` | 2026-03-03 | Low | ## 🎉 Loom v0.1.24  ### 📊 Statistics - **Commits since v0.1.23**: 3 - **Test Status**: ⚠️ Some tests failed (see test output) - **Linters**: ✅ Passed (Go, JavaScript, YAML, Docs, API validation)  ### 📝 Changes  - test: update bead fixture snapshots from test run - fix: DefaultConfig HTTP port 8081 → 8080 to match actual server binding - docs: Update CHANGELOG for v0.1.23 release  ### 🔗 Links - [Full Changelog](https://github.com/jordanhubbard/Loom/compare/v0.1.23...v0.1.24) - [Documentation]( |
| `v0.1.23` | 2026-03-03 | Low | ## 🎉 Loom v0.1.23  ### 📊 Statistics - **Commits since v0.1.22**: 82 - **Test Status**: ⚠️ Some tests failed (see test output) - **Linters**: ✅ Passed (Go, JavaScript, YAML, Docs, API validation)  ### 📝 Changes  - build: add lint and lint-install targets (go vet) - fix: remove broken mockApp from conversation tests; use nil app instead - fix: goroutine leaks in Ralph loop and motivation engine; propagate context in action handlers - harden: remove agent artifacts, fix code holes, add commit gu |
| `v0.1.21` | 2026-02-28 | Low | ## 🎉 Loom v0.1.21  ### 📊 Statistics - **Commits since v0.1.20**: 5 - **Test Status**: ⚠️ Some tests failed (see test output) - **Linters**: ✅ Passed (Go, JavaScript, YAML, Docs, API validation)  ### 📝 Changes  - chore: update test fixture timestamps from test runs - fix: Makefile health-check and URLs use wrong port - fix: close 5 critical executor design flaws - feat: full agent autonomy and CEO experience fixes - docs: Update CHANGELOG for v0.1.20 release  ### 🔗 Links - [Full Changelog](ht |
| `v0.1.20` | 2026-02-27 | Low | ## 🎉 Loom v0.1.20  ### 📊 Statistics - **Commits since v0.1.19**: 5 - **Test Status**: ⚠️ Some tests failed (see test output) - **Linters**: ✅ Passed (Go, JavaScript, YAML, Docs, API validation)  ### 📝 Changes  - fix: silence spurious docker compose env var warnings - fix: resolve CI lint failures and openclaw data race - fix: repair pre-existing test and vet failures across the tree - feat: add CI/CD monitor (CIMon) that auto-files P0 beads for GitHub Actions failures - fix: clear conversatio |
| `v0.1.19` | 2026-02-26 | Low | Additional agent-generated build error fixes: duplicate function body in loomctl, orphaned code in health watchdog, unused import in connectors-service, agent-created fix_loom.go causing main redeclaration. |
| `v0.1.18` | 2026-02-26 | Low | ## What's New in v0.1.18  ### Agent Architecture Awareness - All agents now receive `LOOM_ARCHITECTURE.md` injected into every bead context — a comprehensive system reference covering bead lifecycle, 6 deadlock patterns with escape strategies, agent roles, execution environment, and key invariants - `LESSONS.md` (per-project accumulated knowledge) also injected into every context - Internal executor tracking fields hidden from agent prompts to reduce noise  ### Bug Fixes - Fixed multiple agent-g |
| `v0.1.17` | 2026-02-26 | Low | ## What's Changed  ### Fixed - **Loop-detected beads now block**: Beads that hit the loop detection threshold (repeated provider errors) are now set to `blocked` instead of bouncing back to `open`, stopping infinite executor recycling - **Agent-generated build errors corrected**:   - Removed duplicate `motivation_state_provider.go` that conflicted with `motivation_provider.go`   - Fixed wrong module path (`loom-project/loom` → `jordanhubbard/loom`)   - Removed broken `motivation.NewEngine()` cal |
| `v0.1.16` | 2026-02-26 | Low | ## What's New  ### Fixed: Providers stuck as unhealthy after restart On restart, providers loaded from the database stayed in their last-known `unhealthy` state forever — there was no mechanism to re-probe them. Now loom immediately re-probes every non-healthy provider at startup via a lightweight chat-completion health check, and also re-probes whenever a provider is updated (e.g. to fix an endpoint or API key).  ### Fixed: New projects had no agents Bootstrapped projects created via `POST /api |

## Citation

- HTML: https://www.freshcrate.ai/projects/loom
- Markdown: https://www.freshcrate.ai/projects/loom.md
- Dependencies JSON: https://www.freshcrate.ai/api/projects/loom/deps

_Generated by freshcrate.ai. Indexes github releases for AI-agent ecosystem packages._