DumiFlex 1d62923923 feat(ai): draw item names, variables and item lists in chat replies like the editors ...

A name the model puts in backticks (`Hair Color` or `[[Hair Color]]`)
is now a chip like any other mention. Variables read the way the editors
draw them: accessors in amber, .neg in red, and the variables the run
sets ($iteration, $model_family and the other WP Model Info ones)
marked read-only. A list of three or more items shows as an item list
with each item's kind, name and category chip.

2026-10-27 04:43:40 +00:00

..

__init__.py

feat(ai): optional AI assistant with Draft with AI for wildcards

2026-10-07 22:13:07 +00:00

test_ai_adult.py

feat(ai): pack rules that choose, a draw relation, picks and model families read off the request (#81)

2026-10-21 14:43:44 +02:00

test_ai_ask.py

fix(ai): ask runs beside page calls, repeat page reads answer at once, scope fix trims (#84)

2026-10-21 22:32:43 +02:00

test_ai_audit_gaps.py

feat(ai): assistant overhaul: tool registry, server-kept chats, search index, security, page tools, web search, saved chats, images, permissions, derivation NOT (schema 10), OpenAI Responses API (#64)

2026-10-14 14:56:12 +02:00

test_ai_budget.py

fix(ai): fit new chats to the model's loaded context window (#66)

2026-10-14 16:44:48 +02:00

test_ai_cards.py

feat(ai): feature cards for small models, Type out replies and live thinking (#101)

2026-10-23 04:57:35 +02:00

test_ai_chat_repeats.py

fix(ai): a card re-sent after the request check's note is told what it lacks (#111)

2026-10-23 22:07:18 +02:00

test_ai_chat_tools.py

chore: fixes from the 2026-10-21 review (#93)

2026-10-22 06:15:13 +02:00

test_ai_chat.py

fix(ai): hold Draft with AI and chat edits to what the request asks (#128)

2026-10-26 18:34:51 +01:00

test_ai_chatlog.py

chore: fixes from the 2026-10-21 review (#93)

2026-10-22 06:15:13 +02:00

test_ai_config.py

fix(ai): keep model-tied AI settings per provider and model (#123)

2026-10-26 10:37:56 +01:00

test_ai_debug_log.py

feat(ai): debug log of every AI request with Copy debug (#122)

2026-10-26 09:56:13 +01:00

test_ai_fits.py

feat(ai): chat round, node negatives, pack builder quality fixes (#69)

2026-10-17 01:23:58 +02:00

test_ai_followup.py

feat(ai): refine an answer with a follow-up instead of starting over

2026-10-12 12:07:49 +00:00

test_ai_guide.py

feat(ai): chat round, node negatives, pack builder quality fixes (#69)

2026-10-17 01:23:58 +02:00

test_ai_index.py

fix(ai-chat): hide leaked step fields, live @ chips, reply times (#97)

2026-10-23 00:49:11 +02:00

test_ai_linking.py

feat(ai): draw item names, variables and item lists in chat replies like the editors

2026-10-27 04:43:40 +00:00

test_ai_lint.py

fix(ai): ask runs beside page calls, repeat page reads answer at once, scope fix trims (#84)

2026-10-21 22:32:43 +02:00

test_ai_manage.py

chore: fixes from the 2026-10-21 review (#93)

2026-10-22 06:15:13 +02:00

test_ai_narrow.py

feat(ai): narrow small models' grammars; questions get no change tools (#102)

2026-10-23 05:49:47 +02:00

test_ai_new_editor.py

chore: fixes from the 2026-10-21 review (#93)

2026-10-22 06:15:13 +02:00

test_ai_organize_tools.py

feat(ai): assistant overhaul: tool registry, server-kept chats, search index, security, page tools, web search, saved chats, images, permissions, derivation NOT (schema 10), OpenAI Responses API (#64)

2026-10-14 14:56:12 +02:00

test_ai_organize.py

feat(ai): chat round, node negatives, pack builder quality fixes (#69)

2026-10-17 01:23:58 +02:00

test_ai_pack_drafts.py

feat(ai): pack plans draft again only when needed; slot options come grouped by tag value (#118)

2026-10-24 21:06:34 +02:00

test_ai_pack_extend.py

feat(ai): pack builder overhaul, extend bundles with AI, images and web search in Draft with AI (#65)

2026-10-14 15:17:29 +02:00

test_ai_pack_graph_change.py

feat(ai): pack rules that choose, a draw relation, picks and model families read off the request (#81)

2026-10-21 14:43:44 +02:00

test_ai_pack_graph.py

feat(ai): recommended pack settings, Open in Pack builder, pack 11/15 and Fixed values fixes (#121)

2026-10-26 09:55:58 +01:00

test_ai_pack_parity.py

fix(ai): Claude walkthrough fixes, chat polish and a search results card (#82)

2026-10-21 22:05:41 +02:00

test_ai_pack_quality.py

fix(ai): Claude walkthrough fixes, chat polish and a search results card (#82)

2026-10-21 22:05:41 +02:00

test_ai_pack_recommend.py

feat(ai): recommended pack settings, Open in Pack builder, pack 11/15 and Fixed values fixes (#121)

2026-10-26 09:55:58 +01:00

test_ai_pack_request_needs.py

fix(ai): fold slots split by their own axis, drop self-appends, check rules the way they run (#120)

2026-10-25 00:53:35 +02:00

test_ai_pack_settings.py

feat(ai): pack rules that choose, a draw relation, picks and model families read off the request (#81)

2026-10-21 14:43:44 +02:00

test_ai_pack_smarter.py

fix(ai): Claude walkthrough fixes, chat polish and a search results card (#82)

2026-10-21 22:05:41 +02:00

test_ai_pack_tidy.py

feat(ai): recommended pack settings, Open in Pack builder, pack 11/15 and Fixed values fixes (#121)

2026-10-26 09:55:58 +01:00

test_ai_pack_v2.py

fix(ai): chat assistant eval and fixes (#79)

2026-10-19 03:15:23 +02:00

test_ai_pack.py

fix(ai): pack builder settings reach the graph build, rules on drawn lists hold, honest feature scores (#115)

2026-10-24 08:26:57 +02:00

test_ai_react.py

chore: fixes from the 2026-10-21 review (#93)

2026-10-22 06:15:13 +02:00

test_ai_request_check.py

fix(ai): after the request check, offer only the fix; mark claims no card backs (#108)

2026-10-23 13:06:01 +02:00

test_ai_request_core.py

fix(ai): exact fix for a ref in braces; renaming a variable brings no list tools (#106)

2026-10-23 10:33:25 +02:00

test_ai_rules.py

feat(ai): recommended pack settings, Open in Pack builder, pack 11/15 and Fixed values fixes (#121)

2026-10-26 09:55:58 +01:00

test_ai_safety_guard.py

fix(ai): read negations and consent in the safety guard, keep ratings in step

2026-10-27 02:48:10 +01:00

test_ai_text_items.py

fix(ai): capped answer strings, long axes grouped, no second colour in front of a colour wildcard (#119)

2026-10-24 23:06:28 +02:00

test_ai_tools.py

feat(chat): small models plan multi-item asks, get a tool shortlist and use-when lines (#90)

2026-10-22 00:56:16 +02:00

test_ai_trace.py

feat(ai): optional samplers for local and OpenAI-compatible servers (#105)

2026-10-23 09:00:47 +02:00

test_ai_use_overrides.py

feat(ai): chat round, node negatives, pack builder quality fixes (#69)

2026-10-17 01:23:58 +02:00

test_ai_vocab.py

feat(ai): pack rules that choose, a draw relation, picks and model families read off the request (#81)

2026-10-21 14:43:44 +02:00

test_ai_wildcard_edit.py

fix(ai): hold Draft with AI and chat edits to what the request asks (#128)

2026-10-26 18:34:51 +01:00

test_ai_wildcard.py

fix(ai): hold Draft with AI and chat edits to what the request asks (#128)

2026-10-26 18:34:51 +01:00

test_asked.py

fix(ai): hold Draft with AI and chat edits to what the request asks (#128)

2026-10-26 18:34:51 +01:00

test_step_fields.py

fix(ai-chat): hide leaked step fields, live @ chips, reply times (#97)

2026-10-23 00:49:11 +02:00