When switching between OpenAI models (e.g., gpt-5-mini to gpt-5.2-codex),
function_call IDs with fc_ prefix trigger pairing validation errors because
OpenAI tracks which fc_xxx IDs were paired with rs_xxx reasoning items.
The fix omits the id field for function_calls from different models, which
avoids the pairing validation while keeping call_id for matching with
function_call_output.
Fixes#886
- Added headers field to base StreamOptions interface
- Updated all providers to merge options.headers with defaults
- Forward headers and onPayload through streamSimple/completeSimple
- Bedrock not supported (uses AWS SDK auth)
Tests were throwing errors instead of skipping on CI where no API keys
are configured. Now uses describe.skipIf() and it.skipIf() patterns
consistent with other tests in the package.
Fixes OpenAI Responses 400 error 'reasoning without following item' by
skipping errored/aborted assistant messages entirely rather than filtering
at the provider level. This covers openai-responses, openai-codex-responses,
and future providers.
Removes strictResponsesPairing compat option (no longer needed).
Closes#838
Split OpenAICompat into OpenAICompletionsCompat and OpenAIResponsesCompat
for type-safe API-specific compat settings. Added strictResponsesPairing
option to suppress orphaned reasoning/tool calls on incomplete turns,
fixing 400 errors on Azure's Responses API which requires strict pairing.
Closes#768
- Replace hardcoded pi->CC tool mappings with single CC tool name list
- Case-insensitive lookup: if tool name matches CC tool, use CC casing
- Remove broken find->Glob mapping (round-trip failed)
- Add test coverage for tool name normalization
When an assistant message has stopReason 'error', its tool calls are now
excluded from pending tool tracking. This prevents synthetic tool results
from being generated for calls that will be dropped by provider-specific
converters (e.g., Codex drops tool calls from errored messages).
Previously, this mismatch caused OpenAI to reject requests with 'No tool
call found for function call output with call_id ...' errors.
fixes#812
Bedrock Claude models require max_tokens to exceed thinking.budget_tokens.
This constraint was handled for anthropic-messages API but missing for
bedrock-converse-stream, causing compaction failures.
Extracted adjustMaxTokensForThinking() helper that:
- Adds thinking budget on top of desired output tokens
- Reduces thinking budget if insufficient room (min 1024 output tokens)
- Applied to both anthropic-messages and bedrock-converse-stream APIs
Instead of skipping unsigned tool calls entirely (which lobotomizes context),
convert them to text with an explicit note telling the model this is historical
context from a different model and not a format to mimic.
This preserves tool call/result context when switching from providers without
thought signatures (e.g. Claude via Antigravity) to Gemini 3.
When 429/500 errors occur during tool execution, empty assistant messages
with stopReason='error' get persisted. These break the tool_use -> tool_result
chain for Claude/Gemini APIs.
Added centralized filtering in transformMessages to skip assistant messages
with empty content and no tool calls. Provider-level filters remain for
defense-in-depth.
Previously, OpenAI-compatible provider settings (like developer role support)
were detected only from the baseUrl. When using custom URLs (e.g., proxies),
detection failed. Now checks model.provider first, then falls back to URL.
Fixes#774
The Editor component now accepts TUI as the first constructor parameter,
enabling it to query terminal dimensions. When content exceeds available
height, the editor scrolls vertically keeping the cursor visible.
Features:
- Max editor height is 30% of terminal rows (minimum 5 lines)
- Page Up/Down keys scroll by page size
- Scroll indicators show lines above/below: ─── ↑ 5 more ───
Breaking change: Editor constructor signature changed from
new Editor(theme)
to
new Editor(tui, theme)
fixes#732
- align Codex Responses provider with Pi static instructions
- simplify Codex request/stream handling and cleanup exports
- keep legacy OpenCode Codex prompt for testing until Pi prompt is allowlisted
