/factors
// the twelve operating rules for agentic software, in four tiers
Pass exact intent and bounded evidence between roles; leave the author transcript behind.
Keep work state, decisions, and learnings in git so the next session reads the repo, not a dead chat.
Give each job the smallest sufficient agent; a one-shot task gets no orchestration.
Bound what an agent can touch and spend: capabilities, budget, steps, attempts.
Research the terrain first, then pave it with instructions, contracts, and tests.
Give each concurrent worker exclusive ownership of what it edits; claims are not locks.
The author supplies evidence; a fresh context judges the exact intent, scope, and result.
Land accepted work with its proof so later changes can test against that floor.
Use Beads for the work trail and a reviewed LLM wiki for reusable, source-linked knowledge.
Build and own the verdict machinery; rent the orchestration around it.
Report ambiguity or failure and stop; the caller decides whether another run exists.
Track what verified work costs and drive it toward the cost of unverified work.
Twelve rules, four tiers, one operating loop, backed by plain files in your repo.