Digest / April 16, 2025
Codex Issues Digest — April 16, 2025
84 new issues filed ·
54 resolved in openai/codex (UTC day).
New issues (84)
- Add MCP support.
- First issue! Just want to say thank you!
- docs: TOC in README
- API Key Security Feature Requests
- "OpenAI rejected the request"
- Supporting Azure OpenAI
- we need baseUrl options
- Openrouter api support
- Installs on macOS but then "zsh: command not found: codex"
- O3 is missing
- npm warns about deprecated packages
- codex crash
- docker build gives an error — `ERROR [4/7] COPY dist/codex.tgz codex.tgz``
- Does it support API key from Azure?
- Support for local / other LLMs
- codex doesn't recognize OPENAI_API_KEY after setting it
- Provide suggestions on model picker
- Codex won't start working - rejection message
- Network error
- Using with ChatGPT Plus/Pro
- NODE_OPTIONS=--no-deprecation: not found
- OpenAI rejects requests
- codex does not respect .env on Mac
- codex fails to launch on macOS due to unsupported env -S shebang
- Codex CLI crashes with Error: spawn rg ENOENT when ripgrep isn’t installed
- Fresh pnpm codex install fails to start
- Crashes out when trying to run git status
- [Error]: The server had an error while processing your request. Sorry about that!
- Rename dangerously-auto-approve-everything to yolo
- High‑level overview of the Codex CLI codebase:
- Why requires an OpenAI API key if it runs in a network-disabled environment?
- OpenAI rejected the request (request ID: req_xxx). Please verify your settings and try again.
- Read API key from codex config
- When trying to work with it, I get the errors and the app crashes.
- Crash when asking a question
- o3 not supported in latest from-source build
- CLI crashes with “server_error: The server had an error while processing your request” when context window is nearly exhausted
- Rename to OpenAI Vibe-CLI
- apply_patch aborted
- rate_limit_exceeded
- throw er; // Unhandled 'error' event
- Extend configuration to accept OpenAI organization and project
- ERROR Bad substitution: DOMPurify.sanitize
- dont carsh in API errors
- [BUG] o3, o3-mini, o4-mini, 4.1 not working
- auto sleep in error `rate_limit_exceeded`
- Add ability to hide sensitive files from agent
- ERROR Bad substitution: subIdx - while editing code in approval mode to full-auto
- ERROR Bad substitution: JSON.stringify
- Rate limit reached for o4-mini
- ERROR Bad substitution
- MCP Server connection
- No tool output found for function call
- extra_forbidden error
- Codex crashes when hitting rate limit rather than waiting
- Does `codex` support Gemini 2.5 Pro or Deepseek?
- Alternative codex implementations
- handle node_modules dir correctly
- Listing all files, including node modules
- vim-like input mode
- "Previous response cannot be used for this organization due to Zero Data Retention"
- How to set reasoning effort?
- Shell Command Explanation
- Does o4-mini use the lower caching pricing when needed? When using codex
- Autocomplete for file paths
- bug: unable to continue after hitting esc esc to interrupt assistant
- feat: don't close program when out of credits or when API key is missing
- How to: Codex CLI Windows Terminal in a PowerShell console
- Implement /cost command and track session usage
- Feat: Support streaming all outputs and also support passing history with -q
- bug: No such file or directory
- bug: Hex code `0x17` doesn't work with some word boundaries
- Add a GitHub Actions–based CI pipeline
- Handle rate limit issues
- feature request: ability to interrupt codex part-way through response.
- anyway to use without putting my credit card info to be verified?
- Graceful backoff on rate limits
- Azure API key
- Create Python version Codex CLI
- Adhere to XDG Base Desktop Specification on Linux platforms
- Adhere to the macOS File System Programming Guide on macOS
- Add support for Ollama
- feat: session persistence
- Can you add MacOS alerts when it's the user's turn?
Resolved issues (54)
- Installs on macOS but then "zsh: command not found: codex"
- we need baseUrl options
- docs: TOC in README
- Codex won't start working - rejection message
- codex doesn't recognize OPENAI_API_KEY after setting it
- First issue! Just want to say thank you!
- Does it support API key from Azure?
- O3 is missing
- Network error
- OpenAI rejects requests
- Why requires an OpenAI API key if it runs in a network-disabled environment?
- High‑level overview of the Codex CLI codebase:
- OpenAI rejected the request (request ID: req_xxx). Please verify your settings and try again.
- Rename dangerously-auto-approve-everything to yolo
- Rename to OpenAI Vibe-CLI
- docker build gives an error — `ERROR [4/7] COPY dist/codex.tgz codex.tgz``
- throw er; // Unhandled 'error' event
- rate_limit_exceeded
- When trying to work with it, I get the errors and the app crashes.
- codex does not respect .env on Mac
- Read API key from codex config
- MCP Server connection
- ERROR Bad substitution
- ERROR Bad substitution: subIdx - while editing code in approval mode to full-auto
- ERROR Bad substitution: DOMPurify.sanitize
- auto sleep in error `rate_limit_exceeded`
- dont carsh in API errors
- Does `codex` support Gemini 2.5 Pro or Deepseek?
- Codex crashes when hitting rate limit rather than waiting
- Listing all files, including node modules
- Alternative codex implementations
- Does o4-mini use the lower caching pricing when needed? When using codex
- Rate limit reached for o4-mini
- o3 not supported in latest from-source build
- NODE_OPTIONS=--no-deprecation: not found
- [BUG] o3, o3-mini, o4-mini, 4.1 not working
- codex fails to launch on macOS due to unsupported env -S shebang
- Handle rate limit issues
- Add a GitHub Actions–based CI pipeline
- How to: Codex CLI Windows Terminal in a PowerShell console
- extra_forbidden error
- ERROR Bad substitution: JSON.stringify
- anyway to use without putting my credit card info to be verified?
- Codex CLI crashes with Error: spawn rg ENOENT when ripgrep isn’t installed
- Fresh pnpm codex install fails to start
- Crash when asking a question
- Crashes out when trying to run git status
- codex crash
- [Error]: The server had an error while processing your request. Sorry about that!
- "OpenAI rejected the request"
- Azure API key
- feature request: ability to interrupt codex part-way through response.
- Create Python version Codex CLI
- Add support for Ollama