smith-explore

Smith Explore — Pre-Change Impact Analysis

A diagnostic workflow that audits the codebase for system-wide impacts BEFORE making changes. Identifies all touch points, surfaces conflicts and redundancies, and produces a structured exploration report. Use as the first phase of smith-new or manually before significant changes.

Arguments: $ARGUMENTS

Vault Logging

Throughout this action, log significant events to the vault session log. Read the session log path from .smith/vault/.current-session. If the file is missing or the vault is not initialized, skip all logging silently.

Append entries using this format:

### [HH:MM:SS] /smith-explore <event>

**User Request:**
> <verbatim user message or feature description that triggered this exploration>

**Synthesized Input:** <brief summary of what's being explored>
**Outcome:** <what was discovered>
**Artifacts:** <files read, patterns found>
**Systems affected:** <system IDs>

Log at these points:

1. On invocation — capture the verbatim exploration request AND the scope
2. After each sub-agent completes — findings summary
3. After conflict detection — conflicts found with severity
4. On completion — total touch points, conflicts requiring resolution

When to Use This

Use /smith-explore when:

• Starting a new feature that may have cross-system impacts
• Making changes to core infrastructure (skills, hooks, config files)
• Uncertain about the blast radius of a proposed change
• Want to understand existing patterns before adding new ones

This skill is automatically invoked as Phase 0 of /smith-new when --explore is passed or when the feature touches core Smith infrastructure.

Phase 1: Scope Detection

Phase 1 now uses the manifest as a map, not a fence — start narrow, expand as needed.

Pre-2026-05: Phase 1 used blind grep/find across the entire codebase. Post-manifest-system: we start with the precomputed manifest (via /smith-navigate) for the common case, falling back to whole-codebase grep when the manifest doesn't cover the query.