gmal-scope-builder/backend/app/services
DJP 09441a20b8 Fix deep extraction max_tokens: 16k→32k, shorter descriptions
Root cause: stop_reason=max_tokens - Claude ran out of output tokens
before finishing the tool call JSON for 50+ assets.

Fix:
- Bump max_tokens from 16000 to 32000 for both normal and deep extraction
- Tell Claude to keep descriptions SHORT (1 sentence max)
- Reduce input data to 35k chars (from 40k) to leave more room for output
- Better stop_reason logging on normal extraction too

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-12 16:51:08 -04:00
..
__init__.py Initial commit - GMAL Scope Builder 2026-03-27 17:35:14 -04:00
ai_descriptions.py AI-enhanced GMAL descriptions + matching fixes 2026-03-28 10:12:04 -04:00
ai_matching.py Custom tier editor, brief analysis feeds matching, exportable questions 2026-04-09 16:02:11 -04:00
doc_parser.py Fix deep extraction max_tokens: 16k→32k, shorter descriptions 2026-04-12 16:51:08 -04:00
excel_parser.py Improve matching, upload UX, collapse fix, full catalog approach 2026-03-27 19:22:08 -04:00
export_excel.py Custom tier editor, brief analysis feeds matching, exportable questions 2026-04-09 16:02:11 -04:00
export_pdf.py Initial commit - GMAL Scope Builder 2026-03-27 17:35:14 -04:00
match_refiner.py P2: Iterative prompting + RFP brief analysis engine 2026-04-09 14:15:31 -04:00
ratecard_builder.py Initial commit - GMAL Scope Builder 2026-03-27 17:35:14 -04:00
rfp_analysis.py P2: Iterative prompting + RFP brief analysis engine 2026-04-09 14:15:31 -04:00
team_shape.py P1: Role-specific efficiency profiles + BTG tool efficiencies 2026-04-09 13:48:30 -04:00
tier_expander.py Client tier mapping + GMAL complexity variant expansion 2026-04-09 15:02:45 -04:00