Reflective tool-use controller

After using a tool, JARVIS checks whether the result was actually verified and revises the plan from receipts.

TESTED , rung 3 of 8 Recorded 22 September 2026. Advanced Systems branch, not merged into production.

Technical

What it is

jarvis/reasoning/judgment.py (jarvis.reasoning.plan-revision/1, tool-use-reflection/1) with an orchestrate.py reflection key.

Truth ladder

Rung 3 of 8: TESTED

Automated tests exercise it in isolation, often against fixtures.

Project record: IMPLEMENTED/TESTED; lane live proof PASS on an isolated core; INTEGRATED false; matrix UNDISPOSITIONED

CONCEPT Designed or specified. No working code yet.
IMPLEMENTED Code exists. Nothing has demonstrated that it behaves correctly.
TESTED Automated tests exercise it in isolation, often against fixtures.
INTEGRATED Wired into the rest of JARVIS and tested across component boundaries.
PRODUCTION-REACHABLE The owner can reach it through the real runtime path, not only through a test harness.
LIVE-PROVEN A bounded probe ran the real code path on a booted JARVIS on the real machine and recorded evidence. The project treats live evidence as expiring after 24 hours, so every live reading here shows the date it was taken.
FORMALLY VERIFIED Its critical property is checked by a proof or exhaustive verification.
RELEASE-QUALIFIED Passed every qualification gate for a named release.

Limits

Lane only, not integrated. 'A dispatch that did not verify stays unverified.' Revisions carry authority NONE and execution_authorized false. Advanced Systems work lives on 2026-09-22 branches, not the production branch; most later commits are marked PENDING_PRIMARY_RELEASE under a measurement hold.

Record

22 September 2026 TESTED , rung 3 of 8 Recorded state: IMPLEMENTED/TESTED; lane live proof PASS on an isolated core; INTEGRATED false; matrix UNDISPOSITIONED

Only states recorded in the public export are listed. Earlier states may exist in the project’s private ledgers.

Connected work

Neighbours