BREAKING CHANGE: Passing a prompt on the command line now starts interactive
mode with the prompt pre-submitted, instead of exiting after completion.
Use --print or -p to get the previous non-interactive behavior.
- Add --print / -p flag for non-interactive mode
- Update runInteractiveMode to accept initial messages
- Update README documentation
- Fix Model Selection Priority docs to include --models scope
PR #47 enhancements:
- Add thinking level syntax to --models (e.g., --models sonnet:high,haiku:low)
- First model in scope used as initial model when starting new session
- Auto-apply thinking level when cycling models with Ctrl+P
- Save both model and thinking to session AND settings for persistence
- Simplify UX by removing autoThinkingDisabled flag
- Fix model matching to prioritize exact matches over partial
- Support provider/modelId format (e.g., openrouter/openai/gpt-5.1-codex)
Issue #45:
- Add --thinking CLI flag to set thinking level directly
- Takes highest priority over all other thinking level sources
Closes#45
Fixes#39
- Added headers field to Model type (provider and model level)
- Model headers override provider headers when merged
- Supported in all APIs:
- Anthropic: defaultHeaders
- OpenAI (completions/responses): defaultHeaders
- Google: httpOptions.headers
- Enables bypassing Cloudflare bot detection for proxied endpoints
- Updated documentation with examples
Also fixed:
- Mistral/Chutes syntax error (iif -> if)
- process.env.ANTHROPIC_API_KEY bug (use delete instead of = undefined)
- Add ~/.pi/agent/models.json config for custom providers (Ollama, vLLM, etc.)
- Support all 4 API types (openai-completions, openai-responses, anthropic-messages, google-generative-ai)
- Live reload models.json on /model selector open
- Smart model defaults per provider (claude-sonnet-4-5, gpt-5.1-codex, etc.)
- Graceful session fallback when saved model missing or no API key
- Validation errors show precise file/field info in CLI and TUI
- Agent knows its own README.md path for self-documentation
- Added gpt-5.1-codex (400k context, 128k output, reasoning)
Fixes#21
- Model selector now only shows models with configured API keys
- Added yellow hint in model selector UI explaining the filtering
- Updated README.md to document the filtering behavior in both API Keys and /model sections
- Bumped version to 0.7.11
- Add /branch slash command to create conversation branches
- New UserMessageSelectorComponent shows all user messages chronologically
- Selecting a message creates new session with messages before selection
- Selected message is placed in editor for modification/resubmission
- SessionManager.createBranchedSession() creates new session files
- Updated README.md and CHANGELOG.md with /branch documentation
- Add Ctrl+W for word deletion (stops at whitespace/punctuation)
- Add Ctrl+U for delete to start of line (merges with previous line at col 0)
- Change Ctrl+K from delete entire line to delete to end of line (merges with next line at EOL)
- Add Option+Backspace support in Ghostty (maps to Ctrl+W via ESC+DEL sequence)
- Cmd+Backspace in Ghostty works as Ctrl+U (terminal sends same control code)
- Update README and CHANGELOG with new keyboard shortcuts
Fixes#2, Fixes#3
- Add optional timeout parameter (in seconds) to bash tool
- No default timeout - commands run until completion unless specified
- Agent can provide timeout when needed for long-running commands
- Update README to reflect optional timeout
- Change title from package name to 'pi'
- Emphasize: radically simple, multi-model support, mid-session switching
- Highlight: CLI for headless tasks, creature comforts
- More concise and compelling
- Reference exa-search repo as concrete example
- Shows web search capability via simple tool pattern
- Demonstrates how easy it is to extend pi with custom tools
- Clear stance: no background bash execution
- Alternative: use tmux or tterminal-cp
- Link to tterminal-cp repo
- Rationale: long-running commands need proper terminal sessions
Planning:
- Clear stance: no built-in planning mode
- Alternative: write plans to PLAN.md files
- Persists across sessions, can be versioned
- Example provided showing structured approach
Auto-compaction:
- Added alternative: switch to bigger context model (Gemini)
- Can summarize session with larger model mid-session
- Clear stance: no built-in to-dos
- Rationale: they confuse models more than help
- Alternative: stateful TODO.md file with checkboxes
- Simple example provided
- Agent can read/update file as needed
- Check if --system-prompt argument is a valid file path
- Load file contents if it exists, otherwise use as literal text
- Works in both interactive and single-shot modes
- Project context and datetime still appended automatically
- Document in README with examples
- Changed package name from @mariozechner/coding-agent to @mariozechner/pi-coding-agent
- Updated README installation command
- Binary name 'pi' remains the same
- Avoids conflict with existing @mariozechner/pi (pods package)
- State upfront: pi does not support MCP
- Provide minimal working example (README + bash script)
- Show simple usage pattern
- Remove verbose explanations
fix(tui): don't show duplicate URL when link text equals href
- Rename 'Available Tools' to 'Tools' with 'Built-in Tools' subsection
- Add 'MCP & Adding Your Own Tools' section
- Explain bash/code approach vs MCP servers
- Highlight token efficiency and composability benefits
- Link to blog post for detailed examples
- Document unrestricted filesystem and command access
- Explain no permission checks or safety rails
- Warn about prompt injection risks
- Provide mitigation suggestions
- Clear about design philosophy and use at own risk
- Context files now appended to system prompt
- Added current date/time to system prompt
- Added current working directory to system prompt
- Date/time and cwd placed at end of system prompt
- Updated README to reflect system prompt integration
- Walk up parent directories to load all AGENT.md/CLAUDE.md files
- Load global context from ~/.pi/agent/AGENT.md or CLAUDE.md
- Load order: global → top-most parent → ... → cwd
- Prefer AGENT.md over CLAUDE.md in each directory
- Each context file injected as separate message
- Updated README with detailed documentation
- text mode: only outputs final assistant message text (default)
- json mode: streams all events as JSON (same as session manager writes)
- rpc mode: JSON output + listens for JSON input on stdin for headless operation
- Suppress informational messages in json/rpc modes
