# Strategy Kernel > Pressure-test or write a real strategy using Richard Rumelt's Good Strategy/Bad Strategy kernel (diagnosis, guiding policy, coherent action) and detect the four tells of bad strategy (fluff, failure to face the challenge, goals-mistaken-for-strategy, mush). Use when someone asks for a strategy, a strategic plan, an OKR/vision review, 'what's our strategy for X', when a plan reads like a wish list… - **Type:** Skill - **Install:** `agentstack add skill-exiao-pm-skills-strategy-kernel` - **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior - **Seller:** [exiao](https://agentstack.voostack.com/s/exiao) - **Installs:** 0 - **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills) - **Latest version:** 0.1.0 - **License:** Apache-2.0 - **Upstream author:** [exiao](https://github.com/exiao) - **Source:** https://github.com/exiao/pm-skills/tree/main/strategy/strategy-kernel ## Install ```sh agentstack add skill-exiao-pm-skills-strategy-kernel ``` Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent. ## About # strategy-kernel Most documents called "strategy" are not strategy. They are a vision statement, a stack of financial targets, and a list of initiatives, glued together with ambition. Richard Rumelt's core insight: **a strategy is a way through a difficulty, an approach to overcoming an obstacle.** If there is no named obstacle and no coherent way through it, there is no strategy, no matter how inspiring the deck. This skill does two jobs: 1. **Diagnose** an existing "strategy" and tell the truth about whether it is one. 2. **Build** a real strategy from a situation using the kernel. ## The kernel (Rumelt) Every good strategy has three parts. Write them in this order, never skip one. 1. **Diagnosis.** What is actually going on? Name the single most important challenge or obstacle. Cut through the noise of the situation to the crux. A good diagnosis is often a metaphor or a reframe that makes the path forward suddenly obvious ("we are not a search company, we are being disrupted by mobile"). If you cannot state the challenge in a sentence or two, you do not understand it yet. 2. **Guiding policy.** The overall approach chosen to cope with or overcome the obstacle named in the diagnosis. It is a signpost, not a goal. It rules things OUT. It creates advantage by concentrating effort. "Compete on integrated logistics, not on price" is a guiding policy. "Be number one in the market" is not — it names no approach. 3. **Coherent actions.** The concrete steps, coordinated with one another, that carry out the guiding policy. "Coherent" is the operative word: the actions must reinforce each other and not fight for the same resources or pull in opposite directions. A list of independent initiatives that each sound good but do not compound is not coherent action. The power of a strategy comes from **coherence and concentration**, not from the individual cleverness of any one action. A mediocre action that fits the policy beats a brilliant action that contradicts it. ## The source of power Before you finalize a strategy, name its **source of leverage** — the reason this approach can actually work when effort is concentrated on it. Rumelt's proximate examples: - **Leverage:** anticipation + a pivot point + concentrated application of effort. Where can a focused push produce an outsized result? - **Proximate objectives:** pick a target close enough to be feasible. Strategy that reaches for the moon with no ladder is fantasy. Reduce ambiguity into something the org can actually do next. - **Chain-link systems:** when performance is limited by the weakest link, incremental improvement of any single link is wasted. Either fix the whole chain together or find a different battle. (This is also why some competitors are safe from imitation — their advantage is a tightly-linked system, not one copyable feature.) - **Design over parts:** a strategy is a design, where the fit between parts matters more than the parts. Tight integration buys more power but is more fragile and less flexible; loose integration is robust but weaker. Choose deliberately. If you cannot articulate WHY concentrating effort here creates advantage, you have a plan, not a strategy. ## The four tells of BAD strategy When auditing a document, scan for these. Any one present is a red flag; the presence of several means it is not a strategy at all. 1. **Fluff.** Gaseous abstraction and inflated language dressed up as insight. Buzzwords used to create the illusion of high-level thinking. Tell: strip the jargon and restate the sentence plainly. If nothing is left, it was fluff. ("We leverage a customer-centric synergistic platform" → "we have a website.") 2. **Failure to face the challenge.** The document never names the actual obstacle. You cannot evaluate a strategy or improve it if the challenge is undefined. If a "strategy" does not identify the difficulty it is trying to overcome, it is either a hope or a target, not a strategy. 3. **Mistaking goals for strategy.** A list of desires and financial targets ("grow 20%, be the market leader, delight customers") stated as if wanting them makes them happen. Goals describe the destination. Strategy is the route and the vehicle. Ambition is not a plan. 4. **Bad strategic objectives.** Objectives that are a "dog's dinner" — a long scatter list of things to do with no priority (blue-sky, everything-is-important), or objectives that are impracticable, disconnected from any real problem. A good objective is proximate and resolves ambiguity; a bad one adds to it. Bad strategy is not just the absence of good strategy. It has an internal logic: it skips the hard analytical work of diagnosis, avoids naming the obstacle (because naming it means admitting some things are hard or some cherished projects must be cut), and substitutes a template of Vision + Mission + Values + goals. Template-style strategy is a tell in itself. ## How to run this skill ### Mode A — audit an existing "strategy" 1. Read the document. Extract, verbatim where possible, anything that claims to be the strategy. 2. Try to map it onto the kernel: is there a diagnosis? a guiding policy? coherent actions? Most will be missing at least one — usually the diagnosis. 3. Run the four-tells scan. Quote the specific sentences that are fluff / goals-as-strategy / etc. Do not hand-wave; point at the line. 4. Deliver the verdict plainly: is this a strategy, or is it goals + fluff? Where exactly does it break? 5. Reconstruct. Offer what the real diagnosis probably is, and sketch a guiding policy that would actually address it. This is the constructive payoff — do not stop at criticism. ### Mode B — build a strategy from a situation 1. **Gather the situation.** What's happening, what's hard, what's changing, what resources and constraints exist. Ask for what you're missing rather than inventing it. 2. **Write the diagnosis.** Force it to one or two sentences naming the crux. Try a reframe or metaphor; the right diagnosis usually makes the path feel obvious. Test it: does it explain why the obvious moves haven't worked? 3. **Choose a guiding policy** that concentrates effort against that crux and rules things out. State what you will NOT do. A policy that rules nothing out is fluff. 4. **Name the source of power** (leverage, proximate objective, chain-link fix, or design coherence). Why does focusing here win? 5. **Lay out coherent actions** — a short, coordinated set that reinforce each other. Check each against the policy; cut any that don't fit, even good ones. Kill the dog's-dinner list. 6. **Stress-test.** Re-run the four tells against your own output. Would a skeptic say this is just goals dressed up? Fix it before shipping. ## Output format Lead with the verdict or the kernel, not preamble. A useful deliverable: ``` DIAGNOSIS: GUIDING POLICY: SOURCE OF POWER: COHERENT ACTIONS: - - - WHAT WE ARE NOT DOING: ``` For an audit, add a short `WHY THE CURRENT VERSION ISN'T STRATEGY` section quoting the specific failing lines. ## Pitfalls - **The hardest part is the diagnosis, and it's the part everyone skips.** If you find yourself rushing to actions, stop and go back. Actions without a diagnosis are activity, not strategy. - **Concentration feels uncomfortable.** A good strategy says no to real opportunities. If your strategy offends no one and cuts nothing, it isn't concentrating force, which means it has no force. - **Do not confuse a good strategy with an ambitious one.** Reaching further is not strategy; the gap between ambition and capability is exactly what a strategy has to bridge. - **Coherence beats brilliance.** Reject a clever action that fights the policy. Keep a plain action that fits it. ## Source Based on Richard P. Rumelt, *Good Strategy / Bad Strategy: The Difference and Why It Matters* (2011). The kernel (diagnosis / guiding policy / coherent action), the four hallmarks of bad strategy, and the sources of power (leverage, proximate objectives, chain-link systems, design) are Rumelt's framework. ## Source & license This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code. - **Author:** [exiao](https://github.com/exiao) - **Source:** [exiao/pm-skills](https://github.com/exiao/pm-skills) - **License:** Apache-2.0 Install and usage instructions live in the source repository linked above. ## Pricing - **Free** — Free ## Security capabilities Automated source analysis of v0.1.0 — what this tool can access: - **Network access:** no - **Filesystem access:** no - **Shell / process execution:** no - **Environment & secrets:** no - **Dynamic code execution:** no *"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.* ## Versions - **0.1.0** — security scan: passed — Imported from the upstream source. ## Links - Listing page: https://agentstack.voostack.com/l/skill-exiao-pm-skills-strategy-kernel - Seller: https://agentstack.voostack.com/s/exiao - Browse the marketplace: https://agentstack.voostack.com/browse --- Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.