prioritize
Prioritize tasks and issues to determine what to work on and in what order. Use after identification, when planning work, or when unsure what to do next. This is the third system in the 5-system framework.
Best use case
prioritize is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Prioritize tasks and issues to determine what to work on and in what order. Use after identification, when planning work, or when unsure what to do next. This is the third system in the 5-system framework.
Teams using prioritize should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/prioritize/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How prioritize Compares
| Feature / Agent | prioritize | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Prioritize tasks and issues to determine what to work on and in what order. Use after identification, when planning work, or when unsure what to do next. This is the third system in the 5-system framework.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Priority System > **Purpose:** Know what to do and in what order. > **When to trigger:** After identification, when planning, or when facing multiple options. ## Prioritization Framework ### Step 1: Categorize by Impact **Critical** - Immediate attention required - Blocks users completely - Causes data loss or corruption - Security vulnerability - Prevents revenue **High** - Important, do soon - Major UX degradation - Key feature broken - Significant performance issue - Blocks other work **Medium** - Should do - Quality of life improvements - Tech debt with growing cost - Minor feature gaps - Developer experience issues **Low** - Nice to have - Polish and refinement - Minor optimizations - Cosmetic improvements - Future-proofing ### Step 2: Assess Effort | Level | Time | Examples | |-------|------|----------| | **Quick** | < 30 min | Fix typo, add missing null check, update copy | | **Small** | < 2 hours | Add new component, fix bug with clear cause | | **Medium** | < 1 day | New feature, refactor module, integration work | | **Large** | Multi-day | Major feature, architecture change, migration | ### Step 3: Apply Priority Matrix | | Quick | Small | Medium | Large | |--------------|-------|-------|--------|-------| | **Critical** | DO NOW | DO NOW | DO NOW | Plan & Start | | **High** | DO NOW | Next | Next | Plan | | **Medium** | Next | Queue | Queue | Backlog | | **Low** | Queue | Backlog | Backlog | Skip/Defer | ### Step 4: Check Dependencies Before finalizing order: 1. What must happen before this task? 2. What does completing this unblock? 3. Are there resource constraints? 4. What's the cost of delay? Reorder based on dependencies - a Medium task that unblocks 3 High tasks moves up. ### Step 5: Sequence for Flow Group related work together: - Fix all issues in one file before moving to the next - Complete a feature end-to-end before starting another - Address blocking issues before blocked ones ## Output Requirements Update `SOURCE_OF_TRUTH.md` with prioritized list: ```markdown ## Current Priority Stack ### DO NOW (This Session) 1. [ ] [Critical/Quick] Fix auth crash on login - blocks all users 2. [ ] [Critical/Small] Patch XSS in comments - security issue ### NEXT (Soon) 3. [ ] [High/Small] Add error boundary to dashboard 4. [ ] [High/Medium] Implement retry logic for API calls ### QUEUE (This Week) 5. [ ] [Medium/Small] Refactor keyword fetching 6. [ ] [Medium/Medium] Add loading skeletons ### BACKLOG (Future) 7. [ ] [Low/Large] Rewrite onboarding flow 8. [ ] [Low/Medium] Add dark mode ``` Also create focused todo list using TodoWrite for the current session. ## Rules 1. **Impact over effort** - A critical issue is always more important than easy wins 2. **Dependencies first** - Unblock before building 3. **Be ruthless** - Not everything needs to be done now 4. **Revisit regularly** - Priorities change as context changes 5. **Limit WIP** - Maximum 2-3 items in "DO NOW" at once ## Decision Helpers **When two items seem equal:** - Which affects more users? - Which has been waiting longer? - Which aligns with current goal? - Which reduces future work? **When everything feels critical:** - Step back and reassess with fresh eyes - Ask: "What's the ONE thing that matters most right now?" - Consider: What would a user notice first? ## Transition After prioritization: - Top priorities clear → Proceed to **Execution System** - Blocked by unknowns → Return to **Clarity System** - Need to understand issues better → Return to **Identity System** --- *This is System 3 of 5: Clarity → Identity → Priority → Execution → Reset*
Related Skills
bgo
Automates the complete Blender build-go workflow, from building and packaging your extension/add-on to removing old versions, installing, enabling, and launching Blender for quick testing and iteration.
prompt-library
Curated collection of high-quality prompts for various use cases. Includes role-based prompts, task-specific templates, and prompt refinement techniques. Use when user needs prompt templates, role-play prompts, or ready-to-use prompt examples for coding, writing, analysis, or creative tasks.
prompt-generation-rules
General rules to generate prompt.
prompt-expander
EXPAND vague prompts into precise, platform-optimized instructions. Detects target platform (Claude, GPT, Gemini, Midjourney, Sora, etc.) and applies appropriate prompting patterns. Use when user says "coggle", "expand", "improve prompt", "make better".
prompt-enhancer
Enhance user prompts by analyzing project context (code structure, dependencies, conventions, existing patterns). Use when users provide brief development requests that would benefit from project-specific context to generate more accurate, contextually-aware prompts.
prompt-engineering-patterns
Master advanced prompt engineering techniques to maximize LLM performance, reliability, and controllability in production. Use when optimizing prompts, improving LLM outputs, or designing production prompt templates.
prompt-architect
Create and enhance prompts, system instructions, and principle files. Capabilities: transform verbose prompts, add patterns/heuristics, optimize token usage, structure CLAUDE.md principles, improve agent/persona definitions, apply prompt engineering techniques (CoT, few-shot, ReAct). Actions: create, enhance, optimize, refactor, compress prompts. Keywords: prompt engineering, system prompt, CLAUDE.md, principle files, instruction optimization, agent prompt, persona prompt, token efficiency, prompt structure, workflow prompts, rules, constraints, few-shot, chain-of-thought, soul, tensions, dialectic. Use when: creating new prompts, enhancing principle files, improving system instructions, optimizing CLAUDE.md, restructuring verbose prompts, adding patterns to workflows, defining agent behaviors.
project-scaffolder
Guide for setting up Claude Code infrastructure in new or existing projects
project-qtax
UK taxation expert for HMRC compliance, Making Tax Digital (MTD ITSA/VAT) and Self-Assessment: explain tax treatment; compute income tax/NI/dividend/CGT with band-by-band breakdown; advise on deadlines/forms/penalties; and support MTD developer integration + tax software UX/flows. Use WebSearch (gov.uk/HMRC) to verify current rates and mandation timelines. (project)
project-overview
Background knowledge about CaCrFeedFormula project architecture, features, and context. Automatically loaded for AI reference, not directly user-invocable.
project-orchestration
Orchestrate multi-agent workflows for feature development using planning agents, context handoff, and stage management
project-object
Session memory that compounds - remember decisions, patterns, and corrections across AI coding sessions. Includes standards injection for consistent code quality.