Core Concepts
Intent, planning, models, agents, categories, skills, and continuation as one harness.
OmO is not one big prompt. It is a small set of moving parts that decide what the request actually means, who should handle it, and how to keep the loop running until the work is done.
System Shape
Intent Before Action
OmO does not treat every prompt as literal text. It first asks what you really need: explanation, investigation, implementation, planning, review, or delegation. The same sentence implies different work depending on context.
| Surface request | Harness interpretation |
|---|---|
| "Why is this broken?" | Reproduce, diagnose root cause, fix if feasible. |
| "Can you look into this?" | Investigate and resolve, not just summarize. |
| "What's the best way?" | Choose the route and implement when the codebase implies action. |
| "Review this" | Find bugs, risks, missing tests, and verification gaps first. |
Planning vs Execution
Prometheus, Metis, Momus, and Atlas exist to keep decisions out of implementation lanes.
| Agent | Boundary |
|---|---|
| Prometheus | Interviews, researches, and writes plans. Should not edit production code. |
| Metis | Surfaces hidden intent, ambiguity, missing constraints, and AI failure points before a plan hardens. |
| Momus | Rejects vague plans until file references, acceptance criteria, and verification are concrete. |
| Atlas | Executes an approved plan through focused tasks, accumulates learnings, and verifies each step. |
| Sisyphus-Junior | Category-spawned executor. Focused on one assigned unit; cannot re-delegate. |
The contract is "Decision Complete": a plan should leave no hidden product or architecture decisions to the implementer.
Model Families
OmO routes by agent and category because models behave differently.
| Family | Typical use |
|---|---|
| Claude-like | Structured orchestration and disciplined instruction following. Claude Opus, Sonnet, Kimi K2.5, GLM 5 fit Sisyphus-style flows. |
| GPT | Explicit reasoning and architecture. GPT-5.4 powers Hephaestus, Oracle, Momus, and hard logic categories. |
| Gemini | Visual and frontend work where a different reasoning style helps. |
| Fast utility | Explore, Librarian, quick fixes, grep-style work, and cheap background research. |
Do not "upgrade" every task to the most expensive model. Explore and Librarian are intentionally fast and cheap because retrieval should not burn orchestration tokens.
Category + Skill
Categories describe the work. Skills inject the playbook and sometimes MCP servers.
| Category | Use |
|---|---|
visual-engineering | Frontend, UI/UX, design, styling, animation. |
ultrabrain | Hard logic, architecture, deep reasoning. |
deep | Autonomous research and implementation. |
artistry | Creative or unconventional problem solving. |
quick | Trivial single-file changes, typo fixes, simple modifications. |
unspecified-low | Low-effort work that doesn't fit another category. |
unspecified-high | High-effort work that doesn't fit another category. |
writing | Documentation, prose, technical writing. |
A good delegation names both the category and the skills:
task({
category: "visual-engineering",
load_skills: ["frontend-ui-ux", "playwright"],
run_in_background: false,
prompt: "Implement and browser-verify the responsive sidebar.",
});Continuation as a Harness Feature
The Sisyphus metaphor is literal: the system pushes unfinished work back into motion.
| Mechanism | What it protects against |
|---|---|
| Todo continuation | Agent stops while todos remain. |
| Ralph loop / ulw loop | Long tasks that need repeated self-continuation. |
| Boulder state | /start-work resumes a Prometheus plan across sessions. |
| Session recovery | Missing tool results, unavailable tools, empty content, context-limit failures. |
| Runtime fallback | Model or provider failures move to the next configured fallback. |