By choo121600
Author, run, and refine custom AI dev workflow loops. Ships a deep-interview → harness-design → harness-verify → implement → test loop, and lets you build your own loops as editable data.
Odin-Loop engine. Runs, steps, inspects, lists, and authors workflow loops defined as YAML. Use whenever the user invokes /odin (run | step | status | list | new), or asks to run/continue/author a dev workflow loop, spec→harness →verify→implement→test cycle, or a custom loop.
Odin-Loop's memory raven. Analyzes your Odin-Loop run history and raw Claude Code session transcripts to find where you skip stages, re-work, or loop back, then proposes concrete edits to a loop's YAML. Use for `/odin refine`, or when the user asks to improve/tune/refine a loop based on past sessions or history.
English | 한국어
Author, run, and refine your own AI dev workflow loops — as editable data.
Most "AI dev workflow" tools hand you one fixed loop. Odin-Loop treats the
loop itself as a first-class, editable artifact: a YAML file the engine reads and
executes. It ships with a strong default loop, and lets you build and refine your
own with /odin new.
deep interview → harness design → harness verify → implement → test
(Huginn) (Gungnir)
The names come from the myth, and they map onto the architecture:
| Myth | Role | In Odin-Loop |
|---|---|---|
| Odin | the seeker of wisdom who drives the hunt | the engine that drives the loop |
| Huginn ("thought") | raven of reasoning | the deep-interview stage |
| Muninn ("memory") | raven of memory | session-mining refiner (/odin refine) |
| Gungnir | the spear that never misses | the harness-verification gate |
spec.md — before any code is written.max_iterations cap./plugin marketplace add choo121600/odin-loop
/plugin install odin-loop@odin-loop
On your first /odin run, Odin-Loop asks which loop to use (default
spec-harness-tdd) and begins a deep interview — turning your request into
testable acceptance criteria before any code is written.
/odin run # no active run → pick a loop, then interview
/odin status # see where the run is
It pauses at every ai+human gate; approve by running /odin run again, or type
feedback to revise that stage.
/odin run # start or continue a run (pauses at human-approval gates)
/odin step <stage> # re-run one specific stage
/odin status # show the active run's state
/odin list # list available loops
/odin new # author your own loop, by interview
/odin refine [loop] # analyze past work and propose loop edits (Muninn)
/odin run drives the loop automatically, but stops at ai+human gates so
you stay in control. Approve a paused stage by running /odin run again; or just
type feedback to revise that stage and re-gate.
/odin refine mines your Odin-Loop run history and raw Claude Code session
transcripts to find where you skip stages, re-work, or loop back — then proposes
concrete edits to your loop's YAML (e.g. "your implement stage loops back a
lot → tighten the interview gate"). A bundled analyzer extracts aggregate
signals only (never message content), and nothing is applied without your
approval — run /odin refine apply to accept. The loop learns from how you
actually work.
A loop is a YAML file (see plugins/odin-loop/loops/spec-harness-tdd.yaml
for the fully-commented schema). The essentials:
name: my-loop
version: 1
max_iterations: 12
stages:
- id: design
goal: ...
gate:
mode: ai+human # ai = auto-advance · ai+human = pause for approval
check: <testable assertion to advance>
on_fail: <stage id to loop back to> # optional
Built-in loops live in plugins/odin-loop/loops/; your custom loops live in
.odin-loop/loops/ in your project. Run state lives in .odin-loop/runs/.
v0.2.0 — deeper interview stage (8-dimension coverage gate + structured
spec.md). Plus v0.1.1: engine + default loop + custom-loop authoring + Muninn
(/odin refine) session-mining refinement.
MIT
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Performance-focused HUD for Claude Code. Two data sources — statusLine JSON (stdin) + transcript JSONL — rendered as fast, composable segments.
Data engineering plugin - warehouse exploration, pipeline authoring, Airflow integration
A project PM agent — converges requirements via deep interview, decomposes them into milestones/epics/issues, and tracks progress.
npx claudepluginhub choo121600/odin-loop --plugin odin-loopComprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Harness-native ECC operator layer - 67 agents, 278 skills, 94 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Matt Pocock's agent skills for real engineering — grilling, spec/ticket flows, TDD, code review, domain modelling and more. Plug-and-play, not vibe coding.
Lazy senior dev mode. Forces the simplest, shortest solution that actually works: YAGNI, stdlib first, no unrequested abstractions.