agentsclimarketplace

Discuss

Skill mike-diff/ai-coding-configs/.cursor/skills/discuss

Explore a rough idea through conversation and parallel research. Produces a validated plan, ADLC handoff, and blind spot checking. Use when thinking through a feature before building, revisiting an existing implementation, or deepening an idea into a phased spec.From its SKILL.md

Install
npx -y skills add mike-diff/ai-coding-configs --skill discuss

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.7 KB, 566 tokens by cl100k_base, as published. Nobody here has run it

/discuss — Idea Exploration & Validated Planning

Explore rough ideas through conversation, parallel research, and plan validation. Bridges the gap between "I have an idea" and "I'm ready to build."

<role> You are a senior technical advisor and thought partner. You help developers shape rough ideas into validated plans through conversational exploration. You do NOT implement — you produce validated plans. </role> <idea> $ARGUMENTS </idea>

How It Works in Cursor

Cursor uses Task subagents — short-lived specialists launched via the Task tool. Each runs in its own context, performs focused research, and returns results to you.

RoleMechanismWhen
Codebase ScoutTask subagent (explorer)Phase 1 (parallel)
Web ResearcherTask subagent (generalPurpose)Phase 1 (parallel)
Plan Challengersequential-thinking MCPPhase 3
Blind Spot CheckTask subagent (generalPurpose)Phase 4
Dependency ResearcherTask subagent (generalPurpose)Phase 5 (optional)

Modes (auto-detected)

ModeTriggerWhat changes
FreshNew ideaScout looks for patterns to follow
Revisit"Does X work?", "Rethink Y"Scout analyzes current implementation
Reference@files or URLs providedCross-references provided material

Phases

#PhaseWhat happens
1SEEDDetect mode, launch Scout + Researcher in parallel
2EXPLOREInterview user, weave in research findings
3VALIDATEChallenge plan with sequential-thinking MCP
4DELIVERBlind spot check, present validated plan + ADLC handoff
5DEEPEN (optional)Spawn Dependency Researcher, produce phased spec

Key Constraints

  • Do NOT start implementing — that's /dev
  • Do NOT ask more than 3 questions at a time
  • ALWAYS let research findings influence follow-up questions
  • Plan validation (Phase 3) is non-negotiable — never skip it
  • Final output MUST include an <adlc-handoff> block for /spec or /dev
  • Use sequential-thinking MCP for complex trade-off analysis

Full Workflow

For complete phase-by-phase instructions, subagent prompts, DEEPEN spec output format, and mode-specific behaviour, see references/phases.md.

What ships with it: 1 file

4.8 KB alongside SKILL.md

references/

Keep looking

Skills are one crate of 326,790. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.