An agent skill that helps you audit pricing.
Know what to charge, and why. Leave with your value metric, the models worth testing, and the roadmap to ship them.
We email you the skill and add you to the newsletter. Unsubscribe any time.
Built from pricing work with these companies, and consultations with hundreds more:
What's under the hood.
It investigates before it interviews. Everything it asks comes from named sources, and it asks the questions you didn’t know to ask yourself.
- 14 advice patternsmined from hundreds of pricing calls
- 17 requirement themesfrom real vendor evaluations
- 42 expert claimsPoyar, Litterst, Rivera, the AI summit
- The model taxonomyfit criteria and the audit interview
- Your repo, your pricing code, a plans table, a connected Stripe
- Your live pricing page and your docs, once you point at them
- Then it says what it found, so you correct a paragraph instead of writing one
“What are you here to do?”
Seven modes, one engine. Orient, read, interview, research, recommend.
- A value metricthe unit your price should follow
- 2–3 ranked modelseach tied to your evidence
- A competitor tableread from live pages
- An interactive roadmapone file you keep
- A requirements docwhat must be true to charge it
Better than the same model on its own.
Independent reviewers scored three real pricing engagements against one fixed rubric. The skill met every criterion. The same model, working unaided on the same engagements, met fewer than half.
Distilled from the best pricing minds in the industry.
The skill carries what these conversations taught us, and it credits them by name when a recommendation leans on their work. Watch any of them.
Pricing conversations with the teams building these products
Five kickoffs, ready to paste.
Each card ships a long, technical prompt that points the skill down one path and primes it to ask the right questions. It never answers for you. Copy one, install below, and paste it in.
Run a pricing audit
What it settles
- A hypothesis for your value metric
- Where your current model leaks revenue or margin
- The two or three models worth testing, with evidence
Map the competitive landscape
What it settles
- A competitor table: models, metrics, credits, gates
- The aggregate read on how your market prices
- The white space nobody in the table owns
Pressure-test a usage-based shift
What it settles
- A readiness read: instrumentation, limits, governance
- The meter your pricing should actually run on
- A staged migration shape, grandfathering included
Price an AI feature
What it settles
- The packaging call: bundled, add-on, or credits
- Margin protection while model costs move
- The controls that keep enterprise buyers comfortable
Write your pricing requirements
What it settles
- What your pricing and entitlements must actually do
- The requirement themes teams miss until implementation
- A build-or-buy document you can hand to any vendor
Installed in a minute, useful for a quarter.
- 01
Get the skill
Drop your email in the header and the skill lands in your inbox: one zip with a SKILL.md, seven reference files, and two templates. Plain markdown you can read line by line before you run it.
- 02
Add it to your agent
Unzip it into your agent’s skills folder:
- Any agent: Unzip into .agents/skills/, the standard path every compatible client reads.
- Claude Code, Cursor, Copilot, Cline, OpenCode: Also read .claude/skills/, if that is where your skills already live.
- Claude: Settings → Capabilities → Skills → upload the zip.
- 03
Open a door
Paste a starter prompt from above, or just say what you need. The skill reads the context it already has, tells you what it found, then leads: interview, research, recommendation, roadmap.
- Seven guided modes: audit, new product, new feature, usage-based shift, competitive research, model choice, requirements doc
- The model framework and Socratic interview from Schematic’s live pricing engagements
- A competitive sweep that reads live pricing pages and marks what it could not verify
- Expert research, credited by name: Kyle Poyar, Rob Litterst, Marcos Rivera, and the Monetizing AI Summit
- A hypothesis for your value metric, the thing customers actually buy
- Two or three pricing models worth testing, each tied to your evidence and credited to its source
- A competitor table and the landscape read: metrics, credits, free tiers, enterprise gates
- An interactive pricing roadmap, or a requirements document you can hand to any vendor
The same frameworks the Schematic team brings to a live audit, in a file you run yourself.
One skill, any agent.
The skill is a folder of plain markdown, built on the open Agent Skills standard. Drop it in and it runs, whichever of these your team uses.
These eight are the ones we have verified against their own docs. 50+ clients support the standard.
Want to talk to a human instead?
The team behind the skill runs free pricing audits. A 10 minute survey, a roadmap in 24 to 48 hours.