This is a big week. Probably getting next OpenAI and Anthro models, Grok 4.7, Mimo, etc. These open source model releases are why I can't take the "slow down" crowd seriously. I pitted older Mimo, qwen, step, gpt-oss, and other models against each other playing games like Werewolf and Sketch.io-like games where I let them talk shit while they played against each other. Mimo was by far pareto frontier of game-playing for the models that were <$0.15/m input tokens on OpenRouter. Qwen was pareto frontier in the shit talking game though. Qwen's hilarious. https://www.tiktok.com/@clankerfights/video/7642862917582425...
I really don't think any of these SOTA labs are doing agentic engineering correctly. Skills are the universal language of all agent harnesses. If you abstract the taste and prescription out of the skills and into guidance docs, then leave the skills as basically just workflow scaffolding, you can build task-specific workflows that work with any harness like Claude Code, Codex, Antigravity, etc. Technically, you only really need 2 skills, work and review, and with these you can build infinitely complex workflows including self-improving loops. I built this out and have been using it for months. It's been extremely nice. https://github.com/DanMcInerney/orchflows
The OP is not really a workflow manager, it’s a workspace manager that facilitates creating controlled environments where your skills can run. Everything you said is compatible with (and complementary of) the OP project.
With that said, I’ll somewhat disagree with you. I’ve been down the path you’re talking about and while it is incredibly flexible and powerful, it became too difficult to maintain, and too inconsistent between workflow runs, and a pretty hefty waste of tokens to use AI on things that could instead be handled by deterministic scripts. I ended up creating an orchestrator for myself that uses skills as the primary way to tell agents how to execute a step in a workflow, but also directly orchestrates running scripts and managing state in a deterministic way rather than leaving it all up to agents.
Overly complex; yaml files, heavy framework. Same mistake as Claude Code's Dynamic Workflows. Why not just use the dehydrated skills as the workflow skeleton and use custom guidance docs to hydrate the skills with taste and preference depending on the domain of the task? Now you can build a library of small workflows that compose into larger workflow, and you can export any workflow as a single skill to be used in other harnesses. For example, I have a code.md. It's really small, just a bit of taste preference. If I'm using it to hydrate orch-work for coding tasks, then maybe I want to create a code.api.md which hydrates for further specificity if the task is about creating APIs. Then when new models come out, I can just delete code.api.md and leave it as code.md for /orch-work to read from within a workflow because newer models won't need as much prescription.
Part of what's happening is this is running on Kubernetes, which is oft described as "Overly complex; yaml files, heavy framework" but has value regardless, as perceived by being an industry standard. All the things you describe are well and good, but do not address how one runs many of them reliably (from an infra stand point)
You really, really need different skills depending on the model.
If you are using Qwen 27B you need very prescriptive skills.
If you are using Astra you usually want very minimal skills (because it will follow them but be unnecessarily constrained in some contexts)
If you are using Fable then it depends - it will take the skills as general guidelines but ignore them a lot more than Astra does. Sometimes this is good, sometimes not at all.
Right. That's what modular guidance documentation is for. You could have astra.code or qwen.code.api. All reusable in different workflows. Prescription doesn't belong in the skill itself.
The problem is that knowledge workers just need the simplest way possible to automate their work. Currently I feel like Anthro is overcomplicating this. If I'm an accountant, I want to open Claude App, describe my workflow for balancing some books, then have Claude design a reusable WORKFLOW that it runs whenever I ask. The workflow only needs 2 primitive skills to do anything: /work, and /review. Compose these together with guidance docs into workflows, then workflows and call other workflows. It's all just callable skills.
/balance-books hey claude here's the books. Go balance them
Claude then uses the guidance docs and various subagents to review the books, send off parallel workers with cheap models, then review with a more expensive model. Done in the repeatable, correct order with independent review every time. As new models come out, your workflow structure stays the same. You just delete some prescription from the guidance docs.
I ended up building a simplified version of this as /self-improve in https://github.com/DanMcInerney/orchflows. History is the state ledger, memory and RSI just cite the history as evidence and can be rewritten. I feel like strong immutable state is the missing piece of the puzzle for most of these memory libraries.
Did you look at the OKF repo from Google? Open Knowledge seems to be a common term these days for similar solutions. I think OKF is more of the protocol for wiki-for-llm while you have more of the bells and whistles
I don't disagree with any of this. It is generated software, and it's not a novel idea. I didn't mean for it to come off like that. It's just solving an itch that I couldn't find a solution to and I'm getting a lot of personal utility out of it. I do have a lot of experience with agentic memory, multi-agent systems and harnesses and wasn't super impressed by the workflow of Fable calling opus subagents so I figured I'd apply best practices to what already exists to make it a teensy bit better and easier to use.
Cheers. Absent explanation, I do think it's reasonable to assume that you stood by the wording/claims of the README when you posted it, but I appreciate the patch you made to the docs.
FWIW, re: best practices, your install script potentially runs `rm -rf` on the user's global skills whose names shadow your project's.
I've worked a lot with MiMo in my project that pits LLMs against each other in games (clankerfights.ai). It is a very very good model for the price. MiniMax I'd say is smarter, but MiMo really touches near pareto frontier.
