strategic-investigation

Launch parallel Explore agents in plan mode to synthesize idiomatic solutions.

9|10|Updated Nov 15, 2025
One-click install
npx skills add https://github.com/jkomoros/community-patterns --skill strategic-investigation
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: strategic-investigation
Source: https://github.com/jkomoros/community-patterns/tree/main/.claude/skills/strategic-investigation
Command: npx skills add https://github.com/jkomoros/community-patterns --skill strategic-investigation

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Proactive recovery using plan mode and subagents. After 1-2 failed attempts, STOP trying variations. Enter plan mode and launch parallel Explore/Plan agents to find idiomatic solutions instead of spinning wheels.

Core Features & Use Cases

  • Stop and Acknowledge: Immediately acknowledge being stuck and commit to a plan.
  • Plan Mode: Enter plan mode to structure your investigation.
  • Parallel Subagents: Launch 2-4 Explore agents in parallel to pursue different angles.
  • Synthesize Findings: Consolidate findings from all agents to identify the idiomatic solution.
  • Execute with Confidence: Implement the chosen approach with a solid rationale.
  • Handle Continued Failure: If the solution fails, repeat the plan-mode cycle or escalate.

Quick Start

Trigger plan mode when stuck, then spawn 2–4 Explore agents in parallel, collect findings, and implement the idiomatic solution.

Frequently Asked Questions about strategic-investigation

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
When should I stop trying quick fixes and use a debugging strategy instead?

Debugging strategy becomes essential after 1-2 failed attempts, when errors remain unclear, multiple valid approaches exist, or quick hacks are tempting. Plan-driven investigation with parallel exploration surfaces idiomatic solutions faster than repeated variations.

How do I use plan mode to investigate unclear errors systematically?

Acknowledge being stuck, enter plan mode to structure investigation, spawn 2-4 Explore agents in parallel with focused questions on different angles, collect their findings with file paths and code snippets, then synthesize an idiomatic solution aligned with framework conventions.

What's the difference between exploring multiple angles in parallel versus trying fixes sequentially?

Parallel exploration with subagents investigates different approaches simultaneously, surfacing concrete findings and trade-offs quickly. Sequential attempts risk spinning wheels on variations of the same failed approach; parallel investigation breaks that loop and finds framework-aligned solutions.

Can I use parallel investigation agents when standard debugging approaches have failed?

Yes. Parallel Explore agents are designed precisely for this: after standard approaches fail, they pursue different investigation angles concurrently, returning concrete findings with file paths and code snippets to identify the idiomatic solution.

How do I know when to repeat the investigation cycle versus escalating?

If the synthesized solution from parallel agents fails, repeat the plan-mode cycle with refined questions or new investigation angles. Escalate if repeated cycles exhaust framework conventions or reveal structural constraints beyond the immediate problem scope.

What should I prepare before launching parallel exploration agents?

Acknowledge being stuck and commit to a plan; have clear error context and attempted approaches documented; formulate 2-4 focused questions targeting different investigation angles; be ready to collect findings and synthesize them into an executable solution.