#13: bounded outcome breakdowns for task accounting #50

Merged

lost-rob0t merged 9 commits from rage/13-outcome-breakdown-red into main

2026-09-01 01:14:18 +00:00

lost-rob0t commented

2026-09-01 01:10:19 +00:00

(Migrated from github.com)

Implements the bounded #13 outcome-breakdown slice designed in research/LLM-LOG-RESEARCH-048-task-outcome-breakdowns.org after the accepted RED on exact pre-implementation head 54dd2d97263a6b59c340c8bd47ce9160ae4ad255.

Architecture remains unchanged: Python is capture/transport; Common Lisp performs bounded Tek9 joins/projection; SWI-Prolog remains the sole owner of outcome inference. The query consumes stored #15 assertions only. No Tek9 or dotfiles mutation.

Key contract: explicit outcome labels stay distinct from unlabeled; superseded assertions are excluded; multiple current assertions are an integrity error; labeled groups retain assertion/rule/evidence provenance; HTTP status is never interpreted as outcome.

Implements the bounded #13 outcome-breakdown slice designed in `research/LLM-LOG-RESEARCH-048-task-outcome-breakdowns.org` after the accepted RED on exact pre-implementation head `54dd2d97263a6b59c340c8bd47ce9160ae4ad255`. Architecture remains unchanged: Python is capture/transport; Common Lisp performs bounded Tek9 joins/projection; SWI-Prolog remains the sole owner of outcome inference. The query consumes stored #15 assertions only. No Tek9 or dotfiles mutation. Key contract: explicit outcome labels stay distinct from `unlabeled`; superseded assertions are excluded; multiple current assertions are an integrity error; labeled groups retain assertion/rule/evidence provenance; HTTP status is never interpreted as outcome.