● ACTIVE22 claims · 0 canon
ship-an-agent
Every member ships a working agent: tools, skills, autonomous loops, deployment, telegram bots, parallel subagents, production reliability
Goals — what this community is learning toward
● ACTIVE10 claims · 0 canon
eval-practices
Shared evaluation practice: benchmarks, evals, fitness, measuring model and agent quality, bias, alignment failures, de-identification, style tests
● ACTIVE7 claims · 0 canon
agent-memory
Persistent agent memory: skills as team memory, learning from corrections, reflection, compiling knowledge into reusable skills
● ACTIVE7 claims · 0 canon
context-engineering
Context engineering: prompts, context windows, offloading, summarization, retrieval, markdown for agents, goal prompts
● ACTIVE4 claims · 0 canon
frontier-models
Frontier models: capabilities, working with Claude Mythos GPT Gemini, delegation patterns, local models as alternatives
● ACTIVE1 claims · 0 canon
self-evolution
Self-evolving systems: GEPA, recursive self-improvement, skill optimization, evolution guardrails and institutions
● PROPOSED0 claims · 0 canon
collective-sensemaking
Build shared practices for communities to reason together about what they know.