Prompt caching on Codex was not reliably reusable through the proxy because repeated chat-completions requests could reach the upstream without the same continuity envelope. In practice this showed up most clearly with OpenCode, where cache reads worked in the reference client but not through CLIProxyAPI, although the root cause is broader than OpenCode itself.
The proxy was breaking continuity in several ways: executor-layer Codex request preparation stripped prompt_cache_retention, chat-completions translation did not preserve that field, continuity headers used a different shape than the working client behavior, and OpenAI-style Codex requests could be sent without a stable prompt_cache_key. When that happened, session_id fell back to a fresh random value per request, so upstream Codex treated repeated requests as unrelated turns instead of as part of the same cacheable context.
This change fixes that by preserving caller-provided prompt_cache_retention on Codex execution paths, preserving prompt_cache_retention when translating OpenAI chat-completions requests to Codex, aligning Codex continuity headers to session_id, and introducing an explicit Codex continuity policy that derives a stable continuity key from the best available signal. The resolution order prefers an explicit prompt_cache_key, then execution session metadata, then an explicit idempotency key, then stable request-affinity metadata, then a stable client-principal hash, and finally a stable auth-ID hash when no better continuity signal exists.
The same continuity key is applied to both prompt_cache_key in the request body and session_id in the request headers so repeated requests reuse the same upstream cache/session identity. The auth manager also keeps auth selection sticky for repeated request sequences, preventing otherwise-equivalent Codex requests from drifting across different upstream auth contexts and accidentally breaking cache reuse.
To keep the implementation maintainable, the continuity resolution and diagnostics are centralized in a dedicated Codex continuity helper instead of being scattered across executor flow code. Regression coverage now verifies retention preservation, continuity-key precedence, stable auth-ID fallback, websocket parity, translator preservation, and auth-affinity behavior. Manual validation confirmed prompt cache reads now occur through CLIProxyAPI when using Codex via OpenCode, and the fix should also benefit other clients that rely on stable repeated Codex request continuity.
When assistant has tool_calls but no text content, the translator
emitted an empty message into the Responses API input array before
function_call items. The API then couldn't match function_call_output
to its function_call by call_id, returning:
No tool output found for function call ...
Only emit assistant messages that have content parts. Tool-call-only
messages now produce function_call items directly.
Added 9 tests for tool calling translation covering single/parallel
calls, multi-turn conversations, name shortening, empty content
edge cases, and call_id integrity.
The OpenAI Chat Completions translator was silently dropping
response.function_call_arguments.delta and
response.function_call_arguments.done Codex SSE events, meaning
tool call arguments were never streamed incrementally to clients.
Add proper handling mirroring the proven Claude translator pattern:
- response.output_item.added: announce tool call (id, name, empty args)
- response.function_call_arguments.delta: stream argument chunks
- response.function_call_arguments.done: emit full args if no deltas
- response.output_item.done: defensive fallback for backward compat
State tracking via HasReceivedArgumentsDelta and HasToolCallAnnounced
ensures no duplicate argument emission and correct behavior for models
like codex-spark that skip delta events entirely.
- Replaced all instances of `bytes.Clone` with direct references to enhance efficiency.
- Simplified payload handling across executors and translators by eliminating unnecessary data duplication.
- Added conditional logic for Codex instruction injection based on configuration.
- Updated role terminology from "user" to "developer" for better alignment with context.
- Pass through non-function tool definitions like web_search
- Translate tool_choice for built-in tools and function tools
- Add regression tests for built-in tool passthrough
Replace the "~<n>" suffix with "_<n>" when generating unique short names in codex translators (Claude, Gemini, OpenAI chat).
This avoids using a special character in identifiers, improving compatibility with downstream APIs while preserving length constraints.
- Added comprehensive instructions for Codex CLI harness, sandboxing, approvals, and editing constraints to `internal/misc/codex_instructions/`.
- Clarified `approval_policy` configurations and scenarios requiring escalated permissions.
- Provided detailed style and structure guidelines for presenting results in the Codex CLI.
- Added `LoggerPlugin` to log usage metrics for observability.
- Introduced a new `Manager` to handle usage record queuing and plugin registration.
- Integrated new usage reporter and detailed metrics parsing into executors, covering providers like OpenAI, Codex, Claude, and Gemini.
- Improved token usage breakdown across streaming and non-streaming responses.
- Added `examples/custom-provider/main.go` showcasing custom executor and translator integration using the SDK.
- Removed redundant debug logs from translator modules to enhance code cleanliness.
- Updated SDK documentation with new usage and advanced examples.
- Expanded the management API with new endpoints, including request logging and GPT-5 Codex features.
- Renamed constants from uppercase to CamelCase for consistency.
- Replaced redundant file-based auth handling logic with the new `util.CountAuthFiles` helper.
- Fixed various error-handling inconsistencies and enhanced robustness in file operations.
- Streamlined auth client reload logic in server and watcher components.
- Applied minor code readability improvements across multiple packages.
- Added `CodexInstructions(modelName string)` function to dynamically select instructions based on the model (e.g., GPT-5 Codex).
- Introduced `gpt_5_instructions.txt` and `gpt_5_codex_instructions.txt` for respective model configurations.
- Updated translators to pass `modelName` and use the new instruction logic.
- Updated all `github.com/luispater/CLIProxyAPI/internal/...` imports to point to `github.com/luispater/CLIProxyAPI/v5/internal/...`.
- Adjusted `go.mod` to specify `module github.com/luispater/CLIProxyAPI/v5`.
- Introduced `FunctionCallIndex` to track and manage function call indices within `ConvertCliToOpenAIParams`.
- Enhanced handling for `response.completed` and `response.output_item.done` data types to support tool call scenarios.
- Improved logic for restoring original tool names and setting function arguments during response parsing.
- Introduced reverse mapping logic for tool names in translators to restore original names when shortened.
- Enhanced error handling by logging API response errors consistently across handlers.
- Refactored request and response loggers to include API error details, improving debugging capabilities.
- Integrated robust tool name shortening and uniqueness mechanisms for OpenAI, Gemini, and Claude requests.
- Improved handler retry logic to properly capture and respond to errors.
- Renamed `openai` packages to `chat_completions` across translator modules.
- Introduced `openai_responses_handlers` with handlers for `/v1/models` and OpenAI-compatible chat completions endpoints.
- Updated constants and registry identifiers for OpenAI response type.
- Simplified request/response conversions and added detailed retry/error handling.
- Added `golang.org/x/crypto` for additional cryptographic functions.
