Simplify: Code Review and Cleanup

Review all changed files for reuse, quality, and efficiency. Fix any issues found.

Phase 0: Gate — Scope & Tools

Determine what changed and what tooling is available before any analysis.

Scope detection

Run scripts/build-review-packet.py (shipped in this skill's scripts/) to resolve scope and pre-bake one packet — diff plus changed-file list — that the three review agents read by path instead of each re-running git. It auto-detects staged vs. branch vs. worktree and prints packet_path in its JSON. Fall back to git diff (or git diff HEAD) for a manual scope. If there are no git changes, review the most recently modified files that the user mentioned or that you edited earlier in this conversation.

Graph-enhanced tooling gate

If the code-review-graph MCP plugin is available, probe and ensure freshness:

1. Call build_or_update_graph_tool(base=<base>) to run an incremental update.
2. Call list_graph_stats_tool() to verify the graph has nodes and check last_updated.
3. If either call fails or the graph is empty → proceed with the git-only path for the remainder of the review.

Do not retry.

When the graph is available, follow references/code-review-graph-integration.md for enhanced analysis at each subsequent step.

Phase 1: Triage

Always (git-based):

• Review the diff to understand what changed and why.

When graph is available (see references/code-review-graph-integration.md Phase 1):

• Run detect_changes_tool(base=<base>) for risk-scored, priority-ordered triage.
• Run get_review_context_tool(base=<base>) for token-efficient structural context.
• Optionally run get_affected_flows_tool (2+ files) or get_architecture_overview_tool (3+ directories).

Pass all available context — diff, and graph triage data if available — to each agent in Phase 2.

Phase 2: Launch Three Review Agents in Parallel

Use the Agent tool to launch all three agents concurrently in a single message. Give each agent the packet path from Phase 0 as its starting point (plus graph context from Phase 1 if available): the packet has the diff and changed files, but it is a floor, not a ceiling — agents still read surrounding code, callers, and existing utilities the diff does not show (Agent 1's reuse search is inherently off-diff). When graph data includes risk scores, instruct agents to review highest-risk items first.

Agent 1: Code Reuse Review

For each change:

1. Search for existing utilities and helpers that could replace newly written code.

Look for similar patterns elsewhere in the codebase — common locations are utility directories, shared modules, and files adjacent to the changed ones. When graph is available: use semantic_search_nodes_tool and query_graph_tool("callers_of") per the integration reference.

1. Flag any new function that duplicates existing functionality.

Suggest the existing function to use instead.

1. Flag any inline logic that could use an existing utility — hand-rolled string manipulation, manual path handling, custom environment checks, ad-hoc type guards, and similar patterns are common candidates.
2. Orphaned code: when functions were deleted or replaced, verify nothing was left unreferenced.

When graph is available: use refactor_tool(mode="dead_code") per the integration reference.

Before recommending deletion of code that looks redundant or dead, apply Chesterton's Fence: check git blame, nearby comments, and tests to understand _why_ it was written. If the original reason still applies (platform quirk, perf workaround, edge case), keep it. "I don't see why this is here" is not the same as "this isn't needed."

Agent 2: Code Quality Review

Review the same changes for hacky patterns:

1. Redundant state: state that duplicates existing state, cached values that could be derived, observers/effects that could be direct calls
2. Parameter sprawl: adding new parameters to a function instead of generalizing or restructuring existing ones
3. Copy-paste with slight variation: near-duplicate code blocks that should be unified with a shared abstraction
4. Leaky abstractions: exposing internal details that should be encapsulated, or breaking existing abstraction boundaries.

When graph is available: use get_impact_radius_tool per the integration reference.

1. Stringly-typed code: using raw strings where constants, enums (string unions), or branded types already exist in the codebase
2. Oversized functions: functions that have grown beyond reasonable length.

When graph is available: use find_large_functions_tool per the integration reference.

1. Broken contracts: changed function signatures, return types, or behavioral contracts that leave callers out of sync.

Grep for callers in the codebase and verify they still match. When graph is available: use query_graph_tool("callers_of") and query_graph_tool("inheritors_of") per the integration reference.

1. Test coverage gaps: behavior changes without corresponding test updates.

Search for test files that exercise the changed functions. When graph is available: use query_graph_tool("tests_for") per the integration reference.

1. Unnecessary comments: comments explaining WHAT the code does (well-named identifiers already do that), narrating the change, or referencing the task/caller — delete; keep only non-obvious WHY (hidden constraints, subtle invariants, workarounds)
2. Poor names: generic placeholders (data, result, tmp, val, item) where a content-bearing name fits; non-standard abbreviations (usr, cfg, evt) where full words read better; misleading names (a get* that mutates, an is* that returns a non-boolean).

Match the codebase's existing naming conventions rather than imposing new ones.

Agent 3: Efficiency Review

Review the same changes for efficiency:

1. Unnecessary work: redundant computations, repeated file reads, duplicate network/API calls, N+1 patterns.

When graph is available: use query_graph_tool("callees_of") per the integration reference.

1. Missed concurrency: independent operations run sequentially when they could run in parallel
2. Hot-path bloat: new blocking work added to startup or per-request/per-render hot paths.

When graph is available: cross-reference with affected flows from Phase 1.

1. Recurring no-op updates: state/store updates inside polling loops, intervals, or event handlers that fire unconditionally — add a change-detection guard so downstream consumers aren't notified when nothing changed.

Also: if a wrapper function takes an updater/reducer callback, verify it honors same-reference returns (or whatever the "no change" signal is) — otherwise callers' early-return no-ops are silently defeated

1. Unnecessary existence checks: pre-checking file/resource existence before operating (TOCTOU anti-pattern) — operate directly and handle the error
2. Memory: unbounded data structures, missing cleanup, event listener leaks
3. Overly broad operations: reading entire files when only a portion is needed, loading all items when filtering for one.

When graph is available: use query_graph_tool("importers_of") per the integration reference.

Phase 3: Fix Issues

Wait for all three agents to complete. Aggregate their findings and fix each issue directly. If a finding is a false positive or not worth addressing, note it and move on — do not argue with the finding, just skip it.

Preserve behavior exactly. Fixes must keep inputs, outputs, side effects, ordering, and error paths identical. If a fix requires editing existing tests to pass, that's a signal you changed behavior — revert and reconsider rather than updating the tests to match.

Stay in scope. Fix what the agents flagged in the changed code; don't drive-by refactor unrelated code. If a fix's blast radius grows beyond the original diff, surface it to the user instead of silently expanding the change.

When done, briefly summarize what was fixed (or confirm the code was already clean).

References

• references/code-review-graph-integration.md — tool dispatch playbook for code-review-graph MCP plugin (availability gate, required + optional tools per phase, tool quick reference).