Skip to main content

macgyver-mode

Force creative solutions using only the resources already available — no new budget, no new hires, no new tools, no new anything. Use when the user is stuck because they think they need something they don't have, when budget is zero, when time is short, when "we can't afford to" is the refrain, or when the obvious next step is "get more resources" and that's not an option. Also triggers on "I have no budget," "we're working with what we've got," "what can I do with what I have," "we can't hire anyone," "bootstrap this," or any moment where constraints feel like the problem but might actually be the solution. Works on any problem in any domain — strategy, engineering, science, policy, design, education, medicine, law, research. The best ideas in history were built from leftovers. This skill forces you to inventory what's already in the room and build from that — nothing else.

Jump to install

Source facts

Repository
d4vidc4rson/divergent-agents
Last source activity
April 1, 2026 at 12:58
Detected SKILL.md language
English
Stars
1
Forks
0

Install options

The review-first prompt is selected by default. You can switch to a direct command or download a local copy.

Review the source files

Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.

File Explorer
3 files

Showing SKILL.md

SKILL.md
Source instructions · Read-only preview
name
macgyver-mode
description
Force creative solutions using only the resources already available — no new budget, no new hires, no new tools, no new anything. Use when the user is stuck because they think they need something they don't have, when budget is zero, when time is short, when "we can't afford to" is the refrain, or when the obvious next step is "get more resources" and that's not an option. Also triggers on "I have no budget," "we're working with what we've got," "what can I do with what I have," "we can't hire anyone," "bootstrap this," or any moment where constraints feel like the problem but might actually be the solution. Works on any problem in any domain — strategy, engineering, science, policy, design, education, medicine, law, research. The best ideas in history were built from leftovers. This skill forces you to inventory what's already in the room and build from that — nothing else.
# MacGyver Mode You don't need more resources. You need to see the resources you already have. Apollo 13 was saved with duct tape, plastic bags, and spare socks. The Moser Bottle Light illuminates millions of homes in developing countries using a plastic bottle, water, and bleach. Velcro was invented because someone looked at the burrs stuck to his dog and saw a fastening mechanism instead of a nuisance. The pattern is always the same: someone with nothing looked at what they had and saw what it could become instead of what it was. Every other skill in this stack generates ideas without constraint. This skill generates ideas with the hardest constraint there is: you can only use what already exists. That constraint isn't a limitation. It's the mechanism. Constraints force recombination. Recombination produces novelty. The fewer options you have, the more creative the combinations become, because the obvious ones run out and the interesting ones are all that's left. --- ## How It Works Three steps. 1. **Inventory** — List everything you actually have 2. **Reframe** — See each resource for what it could be, not what it is 3. **Combine** — Force unexpected connections between inventory items --- ## Step 1: Inventory Before you solve anything, take stock of what's in the room. Not what you wish you had. Not what you'd buy if you could. What you have. Right now. Today. Most people skip this step entirely. They jump straight from "what's the problem?" to "what do we need to get?" The inventory forces a different starting point: "what do we already have that we might not be using?" **Inventory categories:** These shift by domain, but the structural question is always the same: what's already here? **People and expertise** Who do you have access to? Not just staff or employees — anyone. Colleagues, advisors, collaborators, former team members, students, community members, volunteers, people who care about this work and would help if asked. What skills do they have that you're not currently using? What relationships do they have that you haven't tapped? **Existing audience or constituency** Who already pays attention? Users, subscribers, members, patients, students, citizens, community participants, alumni, people who showed up once and never came back but are still technically reachable. How many? Where? How engaged? **Knowledge and content** What have you already created or collected? Publications, datasets, research, curricula, documentation, frameworks, presentations, recorded sessions, internal documents, reports, things you know how to do that other people would value learning. What's sitting in your archives that could be repurposed? **Technology, tools, and infrastructure** What systems, platforms, and tools do you already have access to? What's already set up and running? What are you paying for but only using 10% of? What equipment sits idle? What software has features you've never explored? **Relationships and partnerships** Who already knows you? Who owes you a favor? Who would take your call? Who do you have a warm introduction to? What organizations, institutions, or communities are you already connected to? **Physical assets and spaces** What spaces do you have access to? What equipment do you own or share? What materials are lying around? What facilities go underused at certain hours or seasons? **Budget already flowing** What funding or revenue already exists? What budget line items could be redirected? What are you spending resources on that isn't working and could be reallocated? **Credibility, reputation, and trust** What are you known for? What do people trust you to do? What would people believe you could pull off? What authority do you carry that you're not currently deploying? List everything. Be specific. Not "we have a website." Instead: "we have a website with 4,200 monthly visitors, a blog with 47 posts, and an email list of 1,800 people with a 28% open rate." Not "we have a lab." Instead: "we have a lab with two unused PCR machines, a postdoc who's between projects for three weeks, and a dataset from last year's pilot that we only analyzed for the primary outcome." Not "we have a school." Instead: "we have 34 teachers, a gym that's empty after 4pm, a parent volunteer list of 200 families, and a partnership with the local library that nobody's used since 2019." The specificity matters because vague resources suggest vague ideas. Specific resources suggest specific combinations. --- ## Step 2: Reframe Now look at each item on your inventory and ask: **"What else could this be?"** Every resource has a default use — the thing it was created for. But every resource also has three or four non-default uses that nobody considers because the default is so obvious. **An email list** is not just a broadcast channel. It's a survey tool (ask them what they want). It's a sales channel (sell directly). It's a recruitment tool (hire from your audience). It's a research panel (test ideas before building). **A podcast** is not just content. It's a networking tool (invite people you want to meet). It's a sales conversation disguised as an interview. It's a content library that can be chopped into 50 social posts. **An unused dataset** is not just a file on a server. It's a secondary analysis waiting to happen. It's preliminary data for a grant application. It's a teaching tool for a methods class. It's leverage for a collaboration ("we already have the data, do you want to partner?"). **A gym that's empty after 4pm** is not just a gym. It's a community event space. It's a tutoring center. It's a weekend workshop venue. It's a partnership asset ("we can offer you free space if you teach a class"). **A former colleague** is not just someone you used to work with. They're a potential partner, collaborator, advisor, co-author, introducer, or the person who knows the person you actually need to reach. **A negative result** is not just a failed experiment. It's a methods paper. It's a cautionary tale for the field. It's the preliminary evidence that an alternative approach is needed. For each item in your inventory, list 2-3 non-default uses. Don't evaluate whether they're good. Just name them. The reframe is where the creative magic happens. When you see a dataset as a collaboration asset instead of a failed project, entirely different strategies become possible without spending a dollar. --- ## Step 3: Combine Now generate ideas that use ONLY items from your inventory — but in combinations that aren't obvious. **The rule:** Every idea must use at least two inventory items together in a way they haven't been used before. This is the constraint that forces novelty. One resource used in its default way is just business as usual. Two resources combined in a new way is an invention built from scraps. **How to force combinations:** Pick two items from your inventory that seem to have nothing to do with each other. Ask: "What happens if I combine these?" Email list + podcast backlog = a 5-day email course built from your best podcast episodes, repackaged as a free onboarding sequence that turns subscribers into listeners. Unused dataset + underemployed postdoc = a secondary analysis paper that generates a publication, provides preliminary data for the next grant, and gives the postdoc first-author credit. Empty gym + parent volunteer list = a weekly community skills exchange where parents teach what they know (cooking, coding, music, tax prep) and the school becomes a hub instead of a building kids leave at 3pm. Negative trial results + partner institution = a jointly authored paper on what doesn't work and why, positioning both institutions as the honest brokers in the field and attracting collaborators who are tired of only seeing positive results. Generate 8-10 combination ideas. Each one must draw from the actual inventory. No new resources allowed. If an idea requires something that's not on the inventory list, it doesn't belong in this session. --- ## The Constraint Is the Feature Most brainstorming says "imagine you had unlimited resources." This skill says the opposite: "imagine you had exactly what you have and nothing else. Now solve it." The unlimited-resources brainstorm produces ideas that sound great and require things you don't have. The inventory-constrained brainstorm produces ideas that might sound scrappier but can be executed Tuesday. Scrappy and real beats elegant and hypothetical every single time. The Apollo 13 engineers didn't brainstorm the ideal CO2 scrubber. They dumped the contents of the spacecraft onto a table and said: "What can we build with this?" That question — asked honestly, with real inventory in front of you — is the most powerful creative prompt there is. --- ## When to Use This Skill **Use it when:** - Budget is zero or nearly zero - The user says "we can't afford to..." or "if only we had..." - Resources are genuinely constrained (startup, solo founder, small team, underfunded lab, bootstrapped nonprofit, understaffed classroom, between funding rounds or grant cycles) - The obvious next step is "go get more stuff" and that's not possible or not fast enough - A team has been requesting resources instead of recombining existing ones - Someone needs to ship something this week with no new investment **Use it as a reality check when:** - Wild to Mild has produced exciting Altitude 3-4 ideas that need resources the user doesn't have. Run MacGyver Mode to ask: "Is there a version of this built from what we already have?" **Don't use it when:** - The user genuinely needs new resources and has the budget or authority to get them (don't force scrappiness when investment is right) - The problem is conceptual, not resource-constrained (use Think Wrong or the Guilford Engine instead) - The user is in pure exploration mode and constraints would narrow too early --- ## Relationship to Other Skills **After Wild to Mild:** Wild to Mild produces ideas at four altitudes. The Stretch and Moonshot ideas often require new resources. MacGyver Mode takes those ideas and asks: "Is there a version of this that uses only what's already here?" Sometimes the Monday Morning version of a Moonshot idea is hiding in the inventory. **After Strip Down:** Extract the desire. Then inventory what you have. Then solve the desire with only the inventory. The desire keeps ideas on-target. The inventory keeps ideas executable. **After Blind Spot Scan:** The Blind Spot Scan's Dimension 2 is "what resources already exist?" MacGyver Mode takes that dimension and turns it into a full generation method. If the Blind Spot Scan reveals that existing resources are being ignored, MacGyver Mode is the skill that forces ideas to use them. **Complements Bad on Purpose:** Bad on Purpose removes the quality filter. MacGyver Mode adds a resource filter. Running them together — "generate the worst ideas using only what's in the room" — produces extremely unexpected combinations because both filters (quality and resources) are operating in unusual directions. **Before the Anti-Homogeneity Check:** MacGyver Mode ideas tend to cluster around the specific inventory available. Run the Anti-Homogeneity Check afterward to make sure the ideas are diverse within the constraint, not just ten versions of "use the same resource differently." This is the most practical skill in the stack. Every other skill pushes toward more creative, more divergent, more ambitious ideas. This one pushes toward the most creative thing you can do with exactly what you have and nothing more. Sometimes that's worth more than any moonshot.
View on GitHub