docs: the Tangle Agent Spine (north-star architecture)#87
Merged
Conversation
…chmarks Canonical north-star: Agent = AgentProfile x runLoop x ExecutionEnvironment. agent-runtime owns the execution kernel (runLoop, loopDispatch, the new ExecutionEnvironment seam in src/loops/); agent-eval (imported downward) provides AgentProfile, RunRecord, drivers, gates, benchmarks. Benchmarks (Harvey, TaxCalcBench) become (TaskSource x ExecutionEnvironment x Scorer) instances our own production agent runs as the agent-under-test — no fork between what we evaluate, self-improve, and ship. Phased build + invariants.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The canonical blueprint for the agent architecture:
Agent = AgentProfile × runLoop × ExecutionEnvironment. agent-runtime owns the execution kernel + the newExecutionEnvironmentseam (src/loops/); agent-eval is imported downward forAgentProfile/RunRecord/drivers/gates/benchmarks. Harvey + TaxCalcBench become environments our own production agent runs as the agent-under-test — no fork between eval, self-improve, and ship. Includes the phased build (Phase 0 portable identity → Phase 1 the seam → Phase 2 benchmark adapters → Phase 3 self-improve over the union → Phase 4 distributed) and the invariants. Placed here (not agent-eval) because the keystone — theExecutionEnvironmentseam +runLoop— lives in this repo.