Competitive Upgrade
Compare the current repo or product against real companies and turn the findings
into practical upgrades for this codebase.
Use this for product/code strategy, not generic market research. Every output must
connect external findings back to concrete repo capabilities, gaps, and buildable
next steps.
When To Use
Use this when the user asks to:
- compare the current project with real companies or live products
- research competitors and improve the repo from what they do well
- find missing features based on market leaders
- benchmark this app against best-in-class tools
- turn competitor/product research into specs, tickets, or a roadmap
- identify 10x feature opportunities grounded in real examples
Routing out:
- Marketing/positioning teardown only ->
competitor-teardown
- Pure ideation without live research ->
game-changing-features
- Output becomes a large ambiguous project ->
wayfinder (if installed)
- Implementation-ready work -> hand off to
to-spec / to-tickets (if installed)
Request template
Compare this project against real competitors and propose upgrades.
This repo/product: <what it is, in one line>
Compare against: <named products or "you pick the best-in-class">
Audience: <who this is for>
Goal: <what I want to improve: retention, onboarding, depth, etc.>
Out of scope: <areas not to touch>
Deliverable: <findings only | findings + specs | findings + tickets>
Follow the competitive-upgrade skill. Research live products rather than
recalling them, cite what you found, and tie every recommendation back to a
concrete capability or gap in this repo. Separate what we should copy from what
we deliberately should not.
Required Behavior
Live company capabilities change. Perform current web research before making any
claim about real products, pricing, features, positioning, public traction, or
recent changes. Never answer competitor questions from memory alone.
Prefer primary or high-quality sources:
- official product pages and docs
- pricing pages
- changelogs and release notes
- GitHub repos (stars, activity, issues)
- launch posts and credible reviews
- app store listings
Do not invent competitor capabilities. Mark uncertainty explicitly and date every
research claim.
Workflow
1. Understand The Current Repo
Inspect the codebase before researching competitors. Determine:
| Area |
Questions |
| Product category |
What kind of app/tool is this? |
| Target user |
Who is it for? |
| Current capabilities |
What can it do today? |
| Core workflow |
What is the main user action? |
| Stack |
What technologies constrain or enable upgrades? |
| Maturity |
Prototype, MVP, production, portfolio, internal tool? |
Output a short current-state summary before recommending anything.
2. Pick Comparison Targets
Choose 3-7 relevant examples. Prioritize:
- direct competitors
- best-in-class tools in the same workflow
- adjacent products with transferable patterns
- open-source equivalents if the repo is developer-facing
- local/regional examples if the product is geography-specific
If the category is unclear, ask the user or propose one with stated caveats.
3. Research Capabilities
For each target gather current evidence:
| Field |
Meaning |
| Product |
Company/tool name |
| Source |
URL or evidence |
| Core promise |
What they claim to solve |
| Key capabilities |
Main user-facing features |
| Workflow patterns |
UX/product patterns worth learning from |
| Integrations |
APIs, exports, auth, platform hooks |
| Pricing/gating |
Free, paid, enterprise-only, usage limits |
| Trust signals |
Security, compliance, reliability, social proof |
| Differentiators |
What makes them stand out |
| Weaknesses |
Complaints, missing pieces, complexity, cost |
Date the research.
4. Build A Capability Matrix
| Capability |
Current Repo |
Company A |
Company B |
Company C |
Upgrade Potential |
| Fast onboarding |
Partial |
Strong |
Strong |
Basic |
High |
| Export/share |
Missing |
PDF/CSV |
Public links |
API |
High |
| Team workflow |
Missing |
Comments |
Roles |
Activity log |
Medium |
Rules:
Current Repo column must come from code inspection, not guesswork.
- Competitor columns must cite sources.
- Not every missing feature is worth building; separate "must match" from
"interesting but irrelevant".
5. Extract Upgrade Opportunities
Group by value and effort:
- Do Now - small, high-value, low-risk changes.
- Do Next - medium features improving the core workflow.
- Explore - large bets, unclear scope, strategy-heavy.
- Do Not Copy - patterns unsuitable for this repo.
Each recommendation includes: Upgrade | Source pattern | Why it matters | Repo fit |
Effort (Low/Medium/High) | Risk | Next artifact (Spec/Ticket/Prototype/Wayfinder map).
Recommend fewer, higher-value upgrades over exhaustive lists.
6. Convert To Buildable Work
| Situation |
Next skill |
| Clear feature |
to-spec |
| Multiple build tasks |
to-tickets |
| Large ambiguous direction |
wayfinder |
| Quick validation needed |
prototype approach directly |
| Clickable plan wanted |
postplan-upload |
| Architecture change needed |
codebase-design |
For substantial plans, produce a clean HTML plan via postplan-upload.
Output Format
# Competitive Upgrade: <Project>
## Current Repo Summary
- Category:
- Target user:
- Current capabilities:
- Main constraints:
## Comparison Targets
| Target | Why included | Sources |
## Capability Matrix
| Capability | Current Repo | Target A | Target B | Target C | Upgrade Potential |
## Key Findings
1. Finding with source-backed evidence (dated).
## Recommended Upgrades
### Do Now / Do Next / Explore
| Upgrade | Inspired by | Why | Effort | Next |
## Do Not Copy
- Pattern: ... Reason: ...
## Proposed Next Step
One concrete recommended action.
Guardrails
- Do not recommend cloning a feature without adapting it to this repo's stack,
users, and maturity.
- No undated or memory-only claims about competitors.
- Do not hide uncertainty; flag gaps in evidence.
- Do not pad the report; fewer high-confidence upgrades beat long lists.
Related Skills
competitor-teardown - marketing/positioning teardown
game-changing-features - 10x ideation without research
to-spec, to-tickets, wayfinder - implementation handoffs (when installed)
postplan-upload - clickable HTML plan for large outputs
1---2name: competitive-upgrade3description: >-4 Compare the current repo, app, or product against real live companies,5 competitors, SaaS tools, and best-in-class examples using current web research,6 then turn findings into repo-specific upgrade ideas, specs, tickets, or plans.7 Use when the user says compare this project with competitors, learn from real8 companies, what features are missing, upgrade this repo based on market research,9 benchmark against live products, product capability gap, make this more like the10 best tools, or find what we should build next. Do NOT use for pure marketing11 teardowns (competitor-teardown) or ideation without research12 (game-changing-features).13license: MIT14metadata:15 source: https://skillmd.com/skills/hongyime/competitive-upgrade16 author: Local setup17 version: "1.0.1"18 platform: "Codex, Antigravity (agy), OpenCode, Muse on Windows, macOS, Linux (Debian)"19---2021# Competitive Upgrade2223Compare the current repo or product against real companies and turn the findings24into practical upgrades for this codebase.2526Use this for product/code strategy, not generic market research. Every output must27connect external findings back to concrete repo capabilities, gaps, and buildable28next steps.2930## When To Use3132Use this when the user asks to:3334- compare the current project with real companies or live products35- research competitors and improve the repo from what they do well36- find missing features based on market leaders37- benchmark this app against best-in-class tools38- turn competitor/product research into specs, tickets, or a roadmap39- identify 10x feature opportunities grounded in real examples4041Routing out:4243- Marketing/positioning teardown only -> `competitor-teardown`44- Pure ideation without live research -> `game-changing-features`45- Output becomes a large ambiguous project -> `wayfinder` (if installed)46- Implementation-ready work -> hand off to `to-spec` / `to-tickets` (if installed)4748## Request template4950```text51Compare this project against real competitors and propose upgrades.5253This repo/product: <what it is, in one line>54Compare against: <named products or "you pick the best-in-class">55Audience: <who this is for>56Goal: <what I want to improve: retention, onboarding, depth, etc.>57Out of scope: <areas not to touch>58Deliverable: <findings only | findings + specs | findings + tickets>5960Follow the competitive-upgrade skill. Research live products rather than61recalling them, cite what you found, and tie every recommendation back to a62concrete capability or gap in this repo. Separate what we should copy from what63we deliberately should not.64```656667## Required Behavior6869Live company capabilities change. Perform current web research before making any70claim about real products, pricing, features, positioning, public traction, or71recent changes. Never answer competitor questions from memory alone.7273Prefer primary or high-quality sources:7475- official product pages and docs76- pricing pages77- changelogs and release notes78- GitHub repos (stars, activity, issues)79- launch posts and credible reviews80- app store listings8182Do not invent competitor capabilities. Mark uncertainty explicitly and date every83research claim.8485## Workflow8687### 1. Understand The Current Repo8889Inspect the codebase before researching competitors. Determine:9091| Area | Questions |92|---|---|93| Product category | What kind of app/tool is this? |94| Target user | Who is it for? |95| Current capabilities | What can it do today? |96| Core workflow | What is the main user action? |97| Stack | What technologies constrain or enable upgrades? |98| Maturity | Prototype, MVP, production, portfolio, internal tool? |99100Output a short current-state summary before recommending anything.101102### 2. Pick Comparison Targets103104Choose 3-7 relevant examples. Prioritize:105106- direct competitors107- best-in-class tools in the same workflow108- adjacent products with transferable patterns109- open-source equivalents if the repo is developer-facing110- local/regional examples if the product is geography-specific111112If the category is unclear, ask the user or propose one with stated caveats.113114### 3. Research Capabilities115116For each target gather current evidence:117118| Field | Meaning |119|---|---|120| Product | Company/tool name |121| Source | URL or evidence |122| Core promise | What they claim to solve |123| Key capabilities | Main user-facing features |124| Workflow patterns | UX/product patterns worth learning from |125| Integrations | APIs, exports, auth, platform hooks |126| Pricing/gating | Free, paid, enterprise-only, usage limits |127| Trust signals | Security, compliance, reliability, social proof |128| Differentiators | What makes them stand out |129| Weaknesses | Complaints, missing pieces, complexity, cost |130131Date the research.132133### 4. Build A Capability Matrix134135| Capability | Current Repo | Company A | Company B | Company C | Upgrade Potential |136|---|---|---|---|---|---|137| Fast onboarding | Partial | Strong | Strong | Basic | High |138| Export/share | Missing | PDF/CSV | Public links | API | High |139| Team workflow | Missing | Comments | Roles | Activity log | Medium |140141Rules:142143- `Current Repo` column must come from code inspection, not guesswork.144- Competitor columns must cite sources.145- Not every missing feature is worth building; separate "must match" from146 "interesting but irrelevant".147148### 5. Extract Upgrade Opportunities149150Group by value and effort:151152- **Do Now** - small, high-value, low-risk changes.153- **Do Next** - medium features improving the core workflow.154- **Explore** - large bets, unclear scope, strategy-heavy.155- **Do Not Copy** - patterns unsuitable for this repo.156157Each recommendation includes: Upgrade | Source pattern | Why it matters | Repo fit |158Effort (Low/Medium/High) | Risk | Next artifact (Spec/Ticket/Prototype/Wayfinder map).159160Recommend fewer, higher-value upgrades over exhaustive lists.161162### 6. Convert To Buildable Work163164| Situation | Next skill |165|---|---|166| Clear feature | `to-spec` |167| Multiple build tasks | `to-tickets` |168| Large ambiguous direction | `wayfinder` |169| Quick validation needed | prototype approach directly |170| Clickable plan wanted | `postplan-upload` |171| Architecture change needed | `codebase-design` |172173For substantial plans, produce a clean HTML plan via `postplan-upload`.174175## Output Format176177```markdown178# Competitive Upgrade: <Project>179180## Current Repo Summary181- Category:182- Target user:183- Current capabilities:184- Main constraints:185186## Comparison Targets187| Target | Why included | Sources |188189## Capability Matrix190| Capability | Current Repo | Target A | Target B | Target C | Upgrade Potential |191192## Key Findings1931. Finding with source-backed evidence (dated).194195## Recommended Upgrades196### Do Now / Do Next / Explore197| Upgrade | Inspired by | Why | Effort | Next |198199## Do Not Copy200- Pattern: ... Reason: ...201202## Proposed Next Step203One concrete recommended action.204```205206## Guardrails207208- Do not recommend cloning a feature without adapting it to this repo's stack,209 users, and maturity.210- No undated or memory-only claims about competitors.211- Do not hide uncertainty; flag gaps in evidence.212- Do not pad the report; fewer high-confidence upgrades beat long lists.213214## Related Skills215216- `competitor-teardown` - marketing/positioning teardown217- `game-changing-features` - 10x ideation without research218- `to-spec`, `to-tickets`, `wayfinder` - implementation handoffs (when installed)219- `postplan-upload` - clickable HTML plan for large outputs