Help us improve
Share bugs, ideas, or general feedback.
From fortran-dev-pipeline
Interactive skill for discussing and designing the next phase of a Fortran scientific project. Facilitates a conversation with the user about goals, scope, and high-level design, producing a markdown plan document as output. Use when the user says "/next-phase-plan", "plan the next phase", "what should the next phase do", "let's figure out the next steps", or wants to decide what the next phase should accomplish before breaking it into tasks. This is the FIRST step in the planning pipeline — its output feeds into /plan-review and then /enrich-phase-plan.
npx claudepluginhub tonywu20/my-claude-marketplace --plugin fortran-dev-pipelineHow this skill is triggered — by the user, by Claude, or both
Slash command
/fortran-dev-pipeline:next-phase-planThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Facilitates a structured discussion to define the next phase of work. Produces a high-level plan document — not a TOML task breakdown. The plan document then goes through `/plan-review` (architectural gate) and `/enrich-phase-plan` (TOML elaboration) before implementation.
Provides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Provides UI/UX resources: 50+ styles, color palettes, font pairings, guidelines, charts for web/mobile across React, Next.js, Vue, Svelte, Tailwind, React Native, Flutter. Aids planning, building, reviewing interfaces.
Share bugs, ideas, or general feedback.
Facilitates a structured discussion to define the next phase of work. Produces a high-level plan document — not a TOML task breakdown. The plan document then goes through /plan-review (architectural gate) and /enrich-phase-plan (TOML elaboration) before implementation.
/next-phase-plan
No arguments. This skill is conversational — it gathers context and discusses with the user.
fd instead of findrg instead of grepAutomatically collect background before engaging the user:
Project memory:
MEMORY_DIR="$HOME/.claude/projects/$(pwd | sd '/' '-')/memory"
Read $MEMORY_DIR/MEMORY.md and every linked memory file.
Recent git history (last 10 commits on main):
git log --oneline -10 main
Existing plan files (if any):
fd -e md -e toml . plans/
Read the most recent plan file to understand what was last planned.
Deferred improvements from prior review rounds:
fd deferred.md notes/pr-reviews/
Read all deferred.md files found — these are improvements the reviewer identified but the plan didn't commission, and they are candidates for this phase.
Execution reports (if any):
fd -e md . execution_reports/
Skim the most recent report to understand what was completed and what failed.
Invoke the fortran-dev-pipeline:fortran-architect agent to synthesize the context and propose a set of goals for the next phase:
You are helping define the next phase of a Fortran scientific project.
<project_memory>
{{MEMORY_CONTENTS}}
</project_memory>
<recent_git_history>
{{GIT_LOG}}
</recent_git_history>
<last_plan>
{{LAST_PLAN_CONTENTS — or "No prior plan found"}}
</last_plan>
<deferred_improvements>
{{DEFERRED_CONTENTS — or "None"}}
</deferred_improvements>
<execution_report>
{{LAST_REPORT_SUMMARY — or "No prior execution report"}}
</execution_report>
Propose candidate goals for the NEXT phase. For each goal:
- State what it achieves and why it's the right next step
- Estimate whether it is a small, medium, or large effort
- Note any dependencies on prior work or on other goals in this list
Also flag any deferred improvements that are now appropriate to incorporate.
Keep the list focused — 3 to 7 goals is ideal. Do not decompose into tasks.
Present the architect's proposal to the user.
Discuss the proposed goals with the user. Typical questions to work through:
Use the fortran-dev-pipeline:fortran-architect agent for follow-up analysis as needed (e.g., if the user wants to explore a specific design direction or numerical algorithm choice).
There is no fixed number of rounds — continue until the user is satisfied with the scope.
Once the scope is agreed, produce a structured markdown plan document and save it to the path the user specifies (default: plans/phase-{N}/PHASE_PLAN.md):
# Phase {N}: {Phase Name}
**Date:** {YYYY-MM-DD}
**Status:** Draft
## Goals
{Numbered list of agreed goals, each with a one-paragraph description of what it achieves and why now.}
## Scope Boundaries
**In scope:**
{Bulleted list of what this phase covers}
**Out of scope:**
{Bulleted list of what is explicitly deferred — prevents scope creep during implementation}
## Design Notes
{Key architectural decisions, numerical algorithm choices, Fortran idioms to use or avoid, and cautions raised during the discussion. This section is the input to the /plan-review gate.}
## Deferred Items Absorbed
{List any deferred improvements from prior phases that this plan incorporates, with a note on where/how they fit. If none, write "None."}
## Open Questions
{Unresolved questions that /plan-review or /enrich-phase-plan may need to address. If none, write "None."}
Commit the plan document:
git add plans/phase-{N}/PHASE_PLAN.md
git commit -m "plan(phase-{N}): initial phase plan — {Phase Name}"
Tell the user:
"Phase {N} plan saved to
plans/phase-{N}/PHASE_PLAN.md.Next steps:
/plan-review plans/phase-{N}/PHASE_PLAN.md— architectural gate; decides on any deferred items and catches design gaps before implementation./enrich-phase-plan plans/phase-{N}/PHASE_PLAN.md— elaborates the plan into an executor-ready TOML task breakdown./compile-planand/implementation-executor— compile and execute."
Will:
Will not:
/enrich-phase-plan's job)/plan-review's job)