Teknium 3628ccc8c4 feat: use 'developer' role for GPT-5 and Codex models (#4498) ...

OpenAI's newer models (GPT-5, Codex) give stronger instruction-following
weight to the 'developer' role vs 'system'. Swap the role at the API
boundary in _build_api_kwargs() for the chat_completions path so internal
message representation stays consistent ('system' everywhere).

Applies regardless of provider — OpenRouter, Nous portal, direct, etc.
The codex_responses path (direct OpenAI) uses 'instructions' instead of
message roles, so it's unaffected.

DEVELOPER_ROLE_MODELS constant in prompt_builder.py defines the matching
model name substrings: ('gpt-5', 'codex').

2026-04-01 14:49:32 -07:00

..

__init__.py

Refactor Terminal and AIAgent cleanup

2026-02-21 22:31:43 -08:00

anthropic_adapter.py

feat(auth): same-provider credential pools with rotation, custom endpoint support, and interactive CLI (#2647)

2026-03-31 03:10:01 -07:00

auxiliary_client.py

feat(auth): same-provider credential pools with rotation, custom endpoint support, and interactive CLI (#2647)

2026-03-31 03:10:01 -07:00

context_compressor.py

fix: cap percentage displays at 100% in stats, gateway, and memory tool (#3599)

2026-03-28 14:55:18 -07:00

context_references.py

fix(security): add .config/gh to read protection for @file references (#4327)

2026-03-31 12:48:30 -07:00

copilot_acp_client.py

fix(acp): preserve leading whitespace in streaming chunks

2026-03-20 09:38:13 -07:00

credential_pool.py

fix: preserve credential_pool through smart routing and defer eager fallback on 429 (#4361)

2026-04-01 01:02:34 -07:00

display.py

feat: add inline diff previews for write actions

2026-04-01 02:13:57 -07:00

insights.py

fix(insights): show cache tokens in overview so total adds up (#4428)

2026-04-01 03:06:47 -07:00

model_metadata.py

feat: add Fireworks context length detection support (#4158)

2026-03-30 20:37:08 -07:00

models_dev.py

feat: add Fireworks context length detection support (#4158)

2026-03-30 20:37:08 -07:00

prompt_builder.py

feat: use 'developer' role for GPT-5 and Codex models (#4498)

2026-04-01 14:49:32 -07:00

prompt_caching.py

fix(prompt-caching): skip top-level cache_control on role:tool for OpenRouter

2026-03-21 16:54:43 -07:00

redact.py

fix(security): add missing GitHub OAuth token patterns and snapshot redact flag

2026-03-31 10:30:48 -07:00

skill_commands.py

feat(skills): support external skill directories via config (#3678)

2026-03-29 00:33:30 -07:00

skill_utils.py

fix(skill_utils): add type check for metadata field in extract_skill_conditions

2026-04-01 11:34:56 -07:00

smart_model_routing.py

fix: preserve credential_pool through smart routing and defer eager fallback on 429 (#4361)

2026-04-01 01:02:34 -07:00

title_generator.py

feat(agent): configurable timeouts for auxiliary LLM calls via config.yaml (#3597)

2026-03-28 14:35:28 -07:00

trajectory.py

Refactor Terminal and AIAgent cleanup

2026-02-21 22:31:43 -08:00

usage_pricing.py

fix: status bar shows 26K instead of 260K for token counts with trailing zeros (#3024)

2026-03-25 12:45:58 -07:00