Ranked by what each change is worth on the composite — the gap to the next rung of the ladder times the dimension’s weight. Computed from the scores, not written by the model; the model’s sentence under each is the action.
Primerkit runs AI agents on market data to do real work and automate actions.
Primerkit lets you test rules against past data and run them on schedule with a local journal.
Use plain terms: run AI rules on market data and automate actions.
All three together: 58 → 69. Every dimension one rung up: 73 (grade B). Ceiling is 95.
Built only from facts already on the page — not the company’s words. Shown so the gap between what was written and what could have been written is concrete rather than a number.
Five alternative headlines in five angles, a subhead, a call-to-action line, and a line-by-line edit for each of the five dimensions — written only from facts already on primerkit.ai, gated like a rating (anything with a struck word is dropped, not sold). Delivered on a private page in about a minute.
Five dimensions on one ladder — 5 · 20 · 35 · 50 · 65 · 80 · 95 — weighted clarity 30%, specificity 20%, jargon 20%, proof 15%, call to action 15%. For each: the rung the page matched, the words that drove it, and what would reach the next rung. Full rubric →
The page states a concrete product concept but uses abstraction (workspace) to describe what it is.
primerkit is a workspace where your AI agent does real work with market data.
Next rungPrimerkit runs AI agents on market data to do real work and automate actions.
No numerical outcomes; some concrete actions are listed but without measurable results.
place a paper order when you say yes.
Next rungPrimerkit lets you test rules against past data and run them on schedule with a local journal.
Contains abstractions like workspace and AI agent; supports readability but adds buzzy terms.
primerkit is a workspace where your AI agent does real work with market data.
Next rungUse plain terms: run AI rules on market data and automate actions.
No named customers, numbers, or verifiable metrics present.
Open source. Nothing here is investment advice: the model never proposes a strategy, and works from the data or from rules you write.
Next rungInclude a specific user or numeric outcome to demonstrate real-world use.
There is a clear action to view the code on GitHub, which is a low-friction, relevant next step.
View on GitHub
Next rungView the project on GitHub to explore the code and docs.
primerkit is a workspace where your AI agent does real work with market data.Used to judge clarity and presence of abstraction.
View on GitHubShows a direct, verifiable CTA/Proof path.
Open source. Nothing here is investment advice: the model never proposes a strategy, and works from the data or from rules you write.Demonstrates openness but not buyer-proof metrics.
primerkit primerkit is a workspace where your AI agent does real work with market data. Ask an AI about markets and you often get a confident paragraph. It is hard to see which numbers it used, hard to test, and nothing happens next. primerkit is built to close that gap, from idea to action. What your agent can do in it: brief you on what moved in rates and markets today; answer from current market data and show you the lines each answer used; compare its answer with what the web says; test a rule you wrote against past data; run that rule each morning on your computer’s schedule; place a paper order when you say yes. You ask in plain words, inside the AI app you already use (Claude Code, Codex, Gemini CLI) or in its own terminal. Every run is kept in a journal on your computer. View on GitHub CONNECTIONS ● model an AI app you already use, a hosted model on your own key, or one on your machine (Ollama or LM Studio) ● PreReason default data source ○ brokers off until you turn one on: /connect FUNCTIONS a code and Enter runs it DG /digest what changed CP /compare briefings and web search BR /brief a briefing as a board JR /journal every run, kept on your machine CH /chart a chart MD /model switch the model CT /catalog every briefing, metric CN /connect data and brokers Open source. Nothing here is investment advice: the model never proposes a strategy, and works from the data or from rules you write.
Strongest: Call to action 80. Weakest: Proof 5. The 2 passes agreed within 4 points. The band 53–63 is where a re-score of this unchanged page would most likely land: on 42 dry re-scores (12 Oct 2026) the median move was 6 points and one in four moved more than 10. A grade only changes when a new score clears the boundary by 5.
Badges are issued to grades A and B. The share card for this page shows the headline with its jargon struck through; paste the URL anywhere that unfurls links.
Scored from the live homepage
on 2026-10-18 by gpt-5-nano,
2 independent passes, median taken.
Page snapshot 1db0e3abab11.
Rubric 7329c240 (current; what it says).
No person reviewed or edited this score. If the rubric and the score disagree, the rubric wins
and the page is re-scored — not hand-corrected.
Re-scored when the homepage text changes.