dynamic-programming-analysis · v1.1.0 · 2026-09-16 · sha256 2da75111a80ef55d
dynamic-programming-analysis v1.1.0A
Immutable. This exact content never changes and is served at /api/v1/blob/2da75111a80ef55d.
No tool to install first
claude plugin marketplace add https://markdownregistry.com/mp/a/art_b6ofhlbmolzhllrt/2da75111a80ef55da49532f6e9798ed460d76420c8e657db9ea77f5ff8c3568d.json
Then /plugin install dynamic-programming-analysis@markdownregistry-b6ofhlbmolzhllrt in Claude Code. It installs this exact version, the one graded A here, pinned to the commit it was graded at, and leaves it there.
--- name: dynamic-programming-analysis description: Break overwhelming problems, decisions, or plans into DP-style sub-problems (base cases, dependencies, reused parts) and solve bottom-up. Use when user says "I don't know where to start", "break this down", "what's the best way to...", or asks for a step-by-step plan, decision matrix, project roadmap, or architecture decomposition. metadata: author: Ronnasayd Machado - github.com/Ronnasayd version: "1.1.0" --- # Dynamic Programming Analysis Breaks any complex problem into DP-style sub-problems: **optimal substructure** (solution depends on smaller solutions) + **overlapping sub-problems** (recurring parts, solve once, reuse). Applies to business, personal decisions, architecture, research, learning, projects. ## Protocol | # | Step | Action | Gate | | --- | ------------------- | ----------------------------------------------------------------------------------- | ----------------------------------------------------- | | 1 | Define main problem | Restate precisely, final state, explicit/implicit constraints | — | | 2 | Decompose | Break into smaller parts; recurse until **base cases** (directly solvable) | tree structure, see `references/response-template.md` | | 3 | Identify overlaps | Which sub-problems recur across branches? Which partial solutions reuse? | mark explicitly — biggest efficiency win | | 4 | Resolution order | Base cases first, then by dependency (prereq before dependent) | execution sequence | | 5 | Solve each | Smallest → largest; compare options if multiple; note which parent problem it feeds | — | | 6 | Compose | Combine sub-solutions into original problem; validate fit | — | | 7 | Verify | Edge cases, external dependencies, optimal vs. good-enough | — | Respond using the template in `references/response-template.md`. ## Adaptations by problem type | Type | Sub-problem | Base case | Overlap | Composition | | ---------------------- | -------------------------- | ------------------------------------ | ------------------------------------------- | ------------------------- | | Decision | each criterion | verifiable facts, not opinions | criteria affecting multiple options | decision table/matrix | | Planning/Project | each deliverable/milestone | atomic task (1 person, 1 day) | shared deps between phases | schedule/roadmap | | Technical/Architecture | each component/module | primitive functions/services | reusable utilities/patterns | integration diagram | | Learning | each prerequisite concept | needs-no-prerequisite knowledge | fundamentals unlocking multiple topics | ordered learning path | | Business/Strategy | each strategic lever | observable metrics, concrete actions | capabilities supporting multiple strategies | integrated strategic plan | ## Quality principles - Granular: each sub-problem has a clear, direct answer. - No pseudo-decomposition: genuinely break down, don't just rename the problem. - Name dependencies explicitly (what depends on what). - Bottom-up only — solving big before small causes rework. - State reuse explicitly when one solution supports another. ## Reference files - `references/response-template.md` — verbatim markdown response template + sub-problem tree text format. - `references/example.md` — worked example (career transition into AI).