Opus 5.5 + Jev: 10 GitHub repos for a full agent stack
A complete agent stack covering routing, planning, execution, memory and security, with branching decisions handed to Jev.
Opus 5.5 + Jev make this AI stack look f…cking illegal 10 GitHub repos for building everything around the model 01 claude-code-router ▸ https://github.com/musistudio/claude-code-router → routes every request to the model you pick, Opus only where it counts 02 litellm ▸ https://github.com/BerriAI/litellm → one gateway in front of 100+ models, spend tracked per call 03 spec-kit ▸ https://github.com/github/spec-kit → spec first, plan second, code last 04 claude-task-master ▸ https://github.com/eyaltoledano/claude-task-master → your PRD becomes tasks the agent works through one by one 05 superpowers ▸ https://github.com/obra/superpowers → a full dev method built from skills: plan, test, then call it done 06 skills ▸ https://github.com/anthropics/skills → Anthropic's own skills, folders Claude loads only when the job needs them 07 graphiti ▸ https://github.com/getzep/graphiti → memory as a temporal knowledge graph, not a dump 08 langgraph ▸ https://github.com/langchain-ai/langgraph → agents as graphs: state, branches, back edges 09 E2B ▸ https://github.com/e2b-dev/E2B → AI-written code runs in isolated cloud sandboxes, not on your machine 10 promptfoo ▸ https://github.com/promptfoo/promptfoo → evals and red teaming before anything ships the architecture: route → spec → tasks → skills → memory → graph → sandbox → evals I'd split the stack like this: route: claude-code-router → litellm plan: spec-kit → claude-task-master work: superpowers → skills state: graphiti → langgraph safety: E2B → promptfoo Opus 5.5 plans and ships. the forks in between, which file, which tool, retry or stop, go to Jev everyone posts the model and the prompt pack the stack around it decides whether your agent is still running next week ⭣
This post quotes another post: x.com/hanakoxbt/status/2101709924828934222