feat: implement enum-based state machine and update changelog #99

tunahorse · 2025-09-11T20:05:11Z

Summary

Implement enum-based state machine for agent completion detection
Update system prompt with new completion marker format
Add configurable tool validation settings
Update changelog with versions 0.0.70-0.0.72
Fix test alignment issues

Changes Made

Agent State Machine: Replaced error-prone boolean flags with robust enum-based state transitions
System Prompt: Updated completion marker from TUNACODE_TASK_COMPLETE to TUNACODE DONE:
Tool Validation: Made validation configurable via tool_strict_validation setting
Documentation: Updated changelog with all recent versions and improvements
Tests: Fixed test cases to match new completion marker format

Test plan

All tests pass
Pre-commit hooks pass
Changelog updated with proper formatting

Summary by CodeRabbit

New Features
- Unified completion signaling with “TUNACODE DONE:”, improving detection in CLI and agent flow.
- Clearer, earlier state/status messages during reasoning and tool execution; more reliable completion after tools.
Documentation
- Updated system rules with a Completion Signaling section.
- Expanded configuration guidance and added tunacode.json example.
- CHANGELOG updated (0.0.70–0.0.72), plus new planning and research docs.
Tests
- Updated tests to validate the new DONE marker and completion behavior.

- Add AgentState enum with 4 states: USER_INPUT, ASSISTANT, TOOL_EXECUTION, RESPONSE - Create AgentStateMachine with thread-safe state transition validation - Implement backward compatibility layer in ResponseState class - Update node_processor.py to use state machine for completion detection - Add transition rules to prevent invalid state changes - Maintain full backward compatibility with existing boolean flags This replaces the error-prone 4-boolean system with a robust state machine that enforces valid transitions and eliminates premature completion issues.

- Add enum-based state machine implementation for v0.0.72 - Document configurable tool validation for v0.0.71 - Include model configuration updates for v0.0.70 - Fix test alignment with new completion marker format - Update system prompt documentation and examples

coderabbitai · 2025-09-11T20:05:19Z

Caution

Review failed

The pull request is closed.

Walkthrough

Introduces an enum-based agent state machine (USER_INPUT, ASSISTANT, TOOL_EXECUTION, RESPONSE), integrates transitions across node processing, updates completion detection to a “TUNACODE DONE:” marker, revises system prompts and agent config for plan mode, adjusts REPL detection, updates tests, adds planning/research docs, and an example config.

Changes

Cohort / File(s)	Summary
State Machine Core `src/tunacode/types.py`, `src/tunacode/core/agents/agent_components/state_transition.py`, `src/tunacode/core/agents/agent_components/response_state.py`	Adds AgentState enum; introduces thread-safe AgentStateMachine with rules; refactors ResponseState to use the state machine with compatibility properties and helpers.
Node Processing Integration `src/tunacode/core/agents/agent_components/node_processor.py`	Wires explicit transitions (ASSISTANT → TOOL_EXECUTION → RESPONSE), replaces boolean completion with state-driven flow and UI logs, ensures RESPONSE after tool handling.
Completion Detection & Prompting `src/tunacode/core/agents/agent_components/task_completion.py`, `src/tunacode/prompts/system.md`, `src/tunacode/core/agents/main.py`, `src/tunacode/core/agents/agent_components/agent_helpers.py`, `src/tunacode/cli/repl.py`, `src/tunacode/core/agents/agent_components/agent_config.py`	Switches completion marker to “TUNACODE DONE:”; regex-based detection; updates system rules and guidance; aligns prompts/help text; REPL anchor check; plan-mode system prompt filtering expanded.
Tests `tests/characterization/test_characterization_main.py`, `tests/test_prompt_changes_validation.py`	Updates expectations to “TUNACODE DONE:” and validates prompt token presence.
Docs & Plans `CHANGELOG.md`, `memory-bank/plan/`, `memory-bank/research/`	Adds CHANGELOG entries for 0.0.70–0.0.72; introduces architecture plans, enum implementation plan, CLI testing framework plan, and research analysis. Removes Unreleased section.
Example Config `tunacode.json.example`	Adds sample configuration with models, env placeholders, settings, ripgrep, tutorial flags, and MCP servers.

Sequence Diagram(s)

sequenceDiagram
  autonumber
  actor User
  participant Agent
  participant StateMachine as AgentStateMachine
  participant Tools
  participant REPL

  User->>Agent: Provide input
  Agent->>StateMachine: transition_to(ASSISTANT)
  Agent->>Agent: Generate response / tool-calls
  alt Tool calls present
    Agent->>StateMachine: transition_to(TOOL_EXECUTION)
    Agent->>Tools: Execute tool(s)
    Tools-->>Agent: Tool results
  end
  Agent->>StateMachine: transition_to(RESPONSE)
  Agent->>Agent: Check for "TUNACODE DONE:" (regex)
  alt DONE detected
    Agent->>StateMachine: set_completion_detected(True)
    Agent-->>User: Final response
  else Not done
    Agent->>StateMachine: transition_to(ASSISTANT)
    Agent-->>User: Follow-up / continue loop
  end
  REPL->>Agent: Monitor output
  REPL->>REPL: Detect anchored DONE marker

Estimated code review effort

🎯 4 (Complex) | ⏱️ ~60 minutes

Possibly related issues

Enhance Agent Loop Architecture: Quality Assessment and Self-Reflection System #94 — Implements the enum-based state machine and completion signaling described here.
Implement Enum-Based State Machine to Replace Binary Completion Logic #98 — Matches added AgentState, state machine rules, response_state refactor, and task_completion updates.

Possibly related PRs

feat/plan mode UI implementation #81 — Also modifies plan-mode system prompts and REPL completion handling, overlapping with agent_config and repl changes.

Poem

I hop through states: USER to ASSISTANT, so keen,
Then TOOL_EXECUTION—gears softly unseen.
I nibble on outputs, RESPONSE in my zone,
“TUNACODE DONE:” and the task is sewn.
Ears up, loops tight, transitions so clean—
A rabbit’s delight in a well-oiled machine. 🐇✨

Tip

👮 Agentic pre-merge checks are now available in preview!

Pro plan users can now enable pre-merge checks in their settings to enforce checklists before merging PRs.

Built-in checks – Quickly apply ready-made checks to enforce title conventions, require pull request descriptions that follow templates, validate linked issues for compliance, and more.
Custom agentic checks – Define your own rules using CodeRabbit’s advanced agentic capabilities to enforce organization-specific policies and workflows. For example, you can instruct CodeRabbit’s agent to verify that API documentation is updated whenever API schema files are modified in a PR. Note: Upto 5 custom checks are currently allowed during the preview period. Pricing for this feature will be announced in a few weeks.

Please see the documentation for more information.

Example:

reviews:
  pre_merge_checks:
    custom_checks:
      - name: "Undocumented Breaking Changes"
        mode: "warning"
        instructions: |
          Pass/fail criteria: All breaking changes to public APIs, CLI flags, environment variables, configuration keys, database schemas, or HTTP/GraphQL endpoints must be documented in the "Breaking Change" section of the PR description and in CHANGELOG.md. Exclude purely internal or private changes (e.g., code not exported from package entry points or explicitly marked as internal).

Please share your feedback with us on this Discord post.

📜 Recent review details

Configuration used: CodeRabbit UI

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 0d19d3c and 07c0d2e.

📒 Files selected for processing (18)

CHANGELOG.md (1 hunks)
memory-bank/plan/2025-09-11_13-30-00_agent_loop_architecture_enhancement.md (1 hunks)
memory-bank/plan/2025-09-11_14-00-00_enum_state_machine_implementation.md (1 hunks)
memory-bank/plan/2025-09-11_14-15-00_automated_cli_tool_testing_framework.md (1 hunks)
memory-bank/research/2025-09-11_13-14-06_agent_loop_enhancement_analysis.md (1 hunks)
src/tunacode/cli/repl.py (2 hunks)
src/tunacode/core/agents/agent_components/agent_config.py (1 hunks)
src/tunacode/core/agents/agent_components/agent_helpers.py (1 hunks)
src/tunacode/core/agents/agent_components/node_processor.py (6 hunks)
src/tunacode/core/agents/agent_components/response_state.py (1 hunks)
src/tunacode/core/agents/agent_components/state_transition.py (1 hunks)
src/tunacode/core/agents/agent_components/task_completion.py (2 hunks)
src/tunacode/core/agents/main.py (4 hunks)
src/tunacode/prompts/system.md (1 hunks)
src/tunacode/types.py (1 hunks)
tests/characterization/test_characterization_main.py (1 hunks)
tests/test_prompt_changes_validation.py (1 hunks)
tunacode.json.example (1 hunks)

✨ Finishing touches

📝 Generate Docstrings

🧪 Generate unit tests

Create PR with unit tests
Post copyable unit tests in a comment
Commit unit tests in branch enhance-agent-loop-architecture

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

sneaek tater added 3 commits September 11, 2025 14:07

Rollback point: pre-changelog commit - Thu Sep 11 14:56:57 CDT 2025

719af0d

tunahorse merged commit d936fbc into master Sep 11, 2025
1 check passed

tunahorse deleted the enhance-agent-loop-architecture branch September 11, 2025 20:05

tunahorse mentioned this pull request Sep 11, 2025

Implement Enum-Based State Machine to Replace Binary Completion Logic #98

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

feat: implement enum-based state machine and update changelog #99

feat: implement enum-based state machine and update changelog #99

Uh oh!

tunahorse commented Sep 11, 2025 •

edited by coderabbitai bot

Loading

Uh oh!

Uh oh!

coderabbitai bot commented Sep 11, 2025 •

edited

Loading

Review failed

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

feat: implement enum-based state machine and update changelog #99

feat: implement enum-based state machine and update changelog #99

Uh oh!

Conversation

tunahorse commented Sep 11, 2025 • edited by coderabbitai bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Changes Made

Test plan

Summary by CodeRabbit

Uh oh!

Uh oh!

coderabbitai bot commented Sep 11, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Review failed

Walkthrough

Changes

Sequence Diagram(s)

Estimated code review effort

Possibly related issues

Possibly related PRs

Poem

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

tunahorse commented Sep 11, 2025 •

edited by coderabbitai bot

Loading

coderabbitai bot commented Sep 11, 2025 •

edited

Loading