スキル一覧に戻る
mway

go-analyzer-dev

by mway

4🍴 0📅 2026年1月22日
GitHubで見るManusで実行

SKILL.md


name: go-analyzer-dev description: Develop, maintain, and audit Go analyzers (go/analysis). Use when creating/updating analyzers, suggested fixes, analysistest golden files, or integration analyzer harnesses. allowed-tools: ["shell", "apply_patch", "read_file", "write_file", "update_plan"] metadata: short-description: Go analyzer development workflow

Go Analyzer Development

Read these reference files before proceeding:

  • references/style.md - Go code style guidelines
  • references/idioms.md - Go conventions and best practices
  • references/concurrency.md - Go thread safety and race conditions
  • references/testing.md - Go test practices
  • references/tooling.md - Go tooling commands
  • references/organization.md - Go code structure and package design
  • references/performance.md - Go optimization guidance
  • ~/.config/agent/core/behavior.md - Critical thinking and evidence requirements
  • ~/.config/agent/core/methodology.md - 5-phase problem-solving framework
  • ~/.config/agent/core/efficiency.md - Parallelization and throughput
  • ~/.config/agent/core/task-management.md - TODO discipline
  • ~/.config/agent/core/principles.md - Engineering principles
  • ~/.config/agent/core/communication.md - Communication standards
  • ~/.config/agent/domain/coding/workflow.md - Universal coding workflow
  • ~/.config/agent/domain/coding/quality.md - Quality priorities
  • ~/.config/agent/domain/coding/safety.md - Runtime and security safety
  • ~/.config/agent/domain/testing/unit.md - Unit testing strategy
  • ~/.config/agent/domain/testing/coverage.md - Coverage expectations

Instructions

Apply all guidance from the reference files listed above.

Use this workflow when building or modifying analyzers:

  1. Analyzer Contract

    • Define purpose, scope, and explicit non-goals.
    • Specify diagnostics, messages, and when fixes are allowed.
    • Document generated-file handling and nolint handling.
  2. AST/Types Strategy

    • Enumerate the AST node kinds you will visit.
    • State required TypesInfo behavior and guard nil/partial info.
    • Prefer inspector with narrow node filters.
  3. Fix Safety Rules

    • Only emit fixes when correctness is provable from local context.
    • Preserve comments/formatting; avoid whole-decl rewrites unless necessary.
    • Ensure edits are minimal, scoped, and stable with token boundaries.
  4. Scenario Matrix

    • Maintain a scenario/edge-case matrix per analyzer.
    • Mark supported/unsupported cases and expected behavior.
    • Map each matrix row to a test (positive or negative).
  5. Testing Strategy

    • Use analysistest for diagnostics and RunWithSuggestedFixes for fixes.
    • Update golden files to reflect expected fixes.
    • Add tests for error-prone contexts (selectors, generics, comments, blank identifiers, control-flow variations).
  6. Integration & Performance

    • Ensure analyzers run together without panics or conflicts.
    • Avoid O(n^2) traversals where possible; reuse computed maps.
  7. Risk & Unknowns

    • Explicitly list policy decisions needed (e.g., generics, selector derefs).
    • Document residual risks and the rationale for unsupported scenarios.

Deliverables (as applicable):

  • Updated analyzer contract (doc or comments)
  • Scenario/edge-case matrix with test mapping
  • Tests (diagnostics + suggested-fix golden files)
  • Integration test coverage confirmation
  • Evidence notes for Verified vs Unverified behavior

Arguments

Target: ${ARGUMENTS}

スコア

総合スコア

40/100

リポジトリの品質指標に基づく評価

SKILL.md

SKILL.mdファイルが含まれている

+20
LICENSE

ライセンスが設定されている

0/10
説明文

100文字以上の説明がある

0/10
人気

GitHub Stars 100以上

0/15
最近の活動

3ヶ月以内に更新がある

0/10
フォーク

10回以上フォークされている

0/5
Issue管理

オープンIssueが50未満

+5
言語

プログラミング言語が設定されている

+5
タグ

1つ以上のタグが設定されている

0/5

レビュー

💬

レビュー機能は近日公開予定です