ci / build (push) Has been cancelled

Details

Add science control APIs and the hypothesis, experiment and promote CLI

Completes FLUID-WP-0006. The loop now runs end to end from the command
line: two competing presentation hypotheses, an experiment that issues a
routing policy rather than touching traffic, an amendment, a stop that
returns traffic to the default, a confirmed outcome, a resolved
competition, and a promotion the gate can refuse.

Starting or stopping an experiment returns the routing policy for the
operator to install rather than installing it. Blueprint 17 keeps the
controller out of the traffic path, and installing from the handler
would put it straight back in; emitting the document keeps the
separation visible instead of implied.

`fluid audit trace` now answers the section 25 questions from events
rather than summary records, and names the rivals a hypothesis beat: an
audit asking which hypotheses were considered is not answered by naming
only the winner.

Two fixes found by driving the CLI rather than only the tests. Go's flag
package stops at the first positional, so ids given after flags silently
swallowed them; ids are now taken before parsing. And there was no way
to attach a revision to the hypothesis that produced it, which left
`audit trace` unable to say why a revision existed -- `hypothesis attach`
closes that.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_014KmVxhJ35tCo7rE7UnLwWu

Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 1116572@bnt-lap001
Assistant-Session: 8ba9bb93-a72a-4883-b189-2499cce5c400

2026-09-04 06:44:47 +02:00

3.4 KiB

Raw Blame History

type

title

domain

repo

status

owner

topic_slug

created

updated

planning_priority

planning_order

depends_on

state_hub_workstream_id

FLUID-WP-0006

workplan

FLUID Science - the adaptation loop closes (Blueprint Phase C)

infotech

fluid-core

done

worsch

fluid-core

2026-09-04

high

FLUID-WP-0005

b7096155-2641-5df2-857b-13d4240e7e3f

FLUID-WP-0006 - FLUID Science

Hypotheses, experiments, fitness and promotion. Entirely human-driven — the Daimon is deliberately absent. Blueprint §50 says the success criterion is proving this loop works cleanly and safely, not autonomous coding.

T01 - Hypothesis records and lifecycle

id: FLUID-WP-0006-T01
status: done
priority: high
state_hub_task_id: "93dd0a6d-3f46-5da5-857d-e0bd49c356db"

Schema doc §2: DRAFT, READY, PRIORITIZED, DESIGNING, EXPERIMENTING, EVALUATING, then ACCEPTED, REJECTED, SUPERSEDED or DEFERRED. Keep observation, explanation, prediction, intervention and result separate — schema doc §18 warns that collapsing them into one narrative destroys criticism and auditability.

T02 - Competition groups

id: FLUID-WP-0006-T02
status: done
priority: medium
state_hub_task_id: "81f6a22f-929e-5c94-bc61-f4e06ad7de16"

Rival explanations of one pressure coexist (schema doc §6). Preserving uncertainty is a feature.

T03 - Hypothesis API

id: FLUID-WP-0006-T03
status: done
priority: medium
state_hub_task_id: "0050d3e4-cc5c-5e1d-9d29-d14bace0e1cf"

Blueprint §44.3: create, compare, prioritize, attach candidate, record outcome.

T04 - Experiment controller

id: FLUID-WP-0006-T04
status: done
priority: high
state_hub_task_id: "36391a67-6a3e-54c9-afc5-181e0b968f85"

Blueprint §16. Declares hypothesis, control, candidate, cohorts, allocation, metrics, guardrails, start and stop conditions, maximum duration. Per §17 it must not touch traffic directly: it writes routing policy the router consumes. Experiments are interruptible.

T05 - Experiment API

id: FLUID-WP-0006-T05
status: done
priority: medium
state_hub_task_id: "2c2a1ff5-f908-565f-ae5b-c0eb469e6482"

Blueprint §44.4: create, start, change bounded allocation, stop, record metrics, finalize.

T06 - Fitness evaluation against hypotheses

id: FLUID-WP-0006-T06
status: done
priority: high
state_hub_task_id: "7223f795-a005-5fe0-9abb-45e4803a3091"

Wire the fitness engine to experiments. Success criteria may not be changed after results are visible without recording the amendment (§18).

T07 - Promotion controller

id: FLUID-WP-0006-T07
status: done
priority: high
state_hub_task_id: "59188ee2-81e6-5331-a0f6-673b7daf871e"

PROMOTE, EXPAND_EXPERIMENT, RETAIN_AS_OPTION, MODIFY, REVERT, ABANDON, DEFER recorded as decisions with an authorizing actor (§19). Human authority only.

T08 - Append-only event log

id: FLUID-WP-0006-T08
status: done
priority: high
state_hub_task_id: "8eb2e90c-6c65-5ef5-be19-d42c2b091d8c"

Every transition above as a fluid_event (schema doc §15). FLUID evolution is itself a system behavior that must remain reconstructable.

T09 - Audit trace command

id: FLUID-WP-0006-T09
status: done
priority: high
state_hub_task_id: "7dd49f06-9cc7-5791-abfb-33b27649ba8a"

fluid audit trace <revision> reconstructs pressure, hypothesis, revision, experiment, fitness and decision, answering all eleven questions in API Standards §25.

3.4 KiB Raw Blame History