I have been disappointed with the relaunch of Fable 5 due to the default switching to Opus 4.8. Over the past year I have been using sporadically initially the GLM models from Z.AI within Factory. When Factory added Missions, it was a game changer for me and my process. I am currently testing an alternative that, so far, is working well for me. Here is the stack: 1. 1. Using ZCode as the harness. Z.ai confirmed that this is the best harness for GLM-5.2. 2. 2. I recreated Factory's Mission Skill. I have been super impressed with their Missions workflow. 3. 3. I have a Abacus.AI account that allows me to use their api for different models. This could be replaced with Openrouter. 4. Here is the sauce: In ZCode I added additional custom models (opus-4-8 and gpt-5.5). In the Mission Skill I have three agents: orchestrator (opus-4-8), worker (glm-5.2) and validator (gpt-5.5). The mission skill builds the detailed long running plans in phases and standsup a Mission Control page to track progress. This has been working well for me and giving me great results. The reason why I am all in on ZCode is that I bought a yearly plan last fall and have a ton compute to use monthly. And since the model is now so much better and closer to the frontier models, I finally think it's time for me to fully test it and give it the best harness and structure to be successful. I added a few screenshots. Thanks to @Ray Fernando for the early Factory Droid and GLM exposure last year. 1.

Fable 5 is back you guys and it's included in the subscription until the 7th of July.. After that you have to pay usage credits to use it.

@Ray Fernando Have you tried GLM-5.2 in their new ZCode app? It's supposed to be the best way to use it as the ZCode harness is built for the model. Also, in the app they're giving 150% of your quota. I have been using it with your parallel-orchestrate skill and it has been promising so far....

Istead of one agent doing everything in a line, the main agent breaks your goal into independent slices and fans them out to a whole team of workers at the same time; some read and analyze your files, some go research the web, and each one comes back with a clean little report. Then the main agent stitches all of it into one result (a roadmap, an audit, a research brief, whatever you asked for). Here is the fun part. I reverse-engineered how a top-tier model decomposed a giant prompt into parallel agents, then packaged that workflow so anyone can run it on demand. And I put it through a real run, bugs included; it read about 2,500 of my own messages, researched the entire tech stack, and handed me back a roadmap in a few minutes with 11 agents going at once. What you get: - A drop-in Cursor skill (unzip into ~/.cursor/skills/, reload, then just describe a big research, analysis, or audit task) - A built-in structured format so every agent reports back the same clean way - A handoff prompt if you want to port the same idea over to Codex This is the kind of thing that used to feel like it needed a team. It does not. The power is already yours; you just have to ask. Drop a comment with what you run it on, and let me cook with you.