feat: named agents, audit timestamps, --trace logging

[kienbui1995]

Named Agents

<!-- .magic-code/agents/reviewer.md -->
---
model: claude-haiku-4-5
description: Code review agent
tools:
- read_file
- grep_search
---
Review code for bugs and style issues.

Auto-discovered, injected into system prompt.

Audit Completeness

Timestamps added to every audit log entry.

--trace

magic-code --trace 'fix the bug'

Shows tool inputs, outputs, and duration at trace level.

183 tests pass (+3 for agents).

Summary by CodeRabbit

Release Notes

• New Features
  • Added --trace CLI flag to enable detailed debug logging of tool executions with timing and output metrics.
  • Introduced support for discovering and loading named agents from project directories, automatically integrating them into system operations.
  • Enhanced audit logging with timestamps for improved tracking and debugging visibility.

[coderabbitai]

Caution

Review failed

Pull request was closed or merged during review

📝 Walkthrough

Walkthrough

The changes introduce agent discovery and loading functionality to the CLI with a new --trace flag for structured logging, enable agents to be discovered from project directories and integrated into system prompts, and enhance observability through trace-level logging of tool execution and audit log timestamps.

Changes

Cohort / File(s)	Summary
CLI Agent Integration mc/crates/mc-cli/src/main.rs	Added --trace boolean flag to CLI struct; updated logging setup to enable trace-level logging for tool registry when flag is set; integrated agent discovery and agents prompt section into system prompt construction.
Agent Module mc/crates/mc-core/src/agents.rs	New module introducing AgentDef struct with name, optional model, description, instructions, and allowed_tools fields. Implements discover_agents() to scan .magic-code/agents/ and agents/ directories for markdown files with YAML frontmatter, parsing agent definitions. Implements agents_prompt_section() to format agents into a markdown section. Includes unit tests for frontmatter parsing and discovery behavior.
Core Module Exports mc/crates/mc-core/src/lib.rs	Added public re-exports of AgentDef, discover_agents, and agents_prompt_section from new agents module.
Observability Enhancements mc/crates/mc-tools/src/audit.rs, mc/crates/mc-tools/src/registry.rs	Added ts (timestamp) field to JSONL audit logs via chrono_now() helper formatting Unix time as HH:MM:SS. Added trace-level logging in ToolRegistry::execute to record tool input before validation and output after execution with elapsed time and output length.

Sequence Diagram

sequenceDiagram
    participant CLI as CLI
    participant Core as mc-core
    participant FS as File System
    participant Prompt as System Prompt
    
    CLI->>CLI: Parse --trace flag
    CLI->>CLI: Configure logging (trace/debug/warn)
    CLI->>Core: discover_agents(project.cwd)
    Core->>FS: Scan .magic-code/agents/ & agents/
    FS-->>Core: Return .md files
    Core->>Core: Parse frontmatter & extract AgentDef
    Core-->>CLI: Return Vec<AgentDef>
    alt Agents found
        CLI->>Core: agents_prompt_section(&agents)
        Core-->>CLI: Formatted agents markdown
        CLI->>Prompt: Append agents section
    else No agents
        CLI->>Core: agents_prompt_section(&agents)
        Core-->>CLI: Empty string
    end
    CLI->>Prompt: Build system prompt with agents

Loading

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~25 minutes

Poem

🐰 ✨ Agents hop into the CLI with traces so bright,
Discovered from directories, configured just right,
Timestamps and logging now chronicle the way,
Tools speak their secrets in a verbose display! 🌙

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 60.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title 'feat: named agents, audit timestamps, --trace logging' accurately and specifically summarizes all three main changes in the PR: agent discovery/support, audit log timestamps, and the new trace logging flag.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch feat/named-agents-audit-trace

---

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