Superdense · Outcome Packs
Set up your first outcome loop
An outcome loop is a small, git-tracked folder where your agent works to improve one real-world outcome. Pick an outcome pack below, get familiar with the core concepts, then follow the four setup steps to start your own.
Outcome packs
Each pack is a ready-made outcome — its goal, levers, and completion gate. Pick one and your agent adapts it to your own account or repo.
North star: impressions per post, measured via X API public_metrics (impression_count). A smaller, right audience tends to beat a larger indifferent one — use your judgment when raw reach trades against relevance for your specific goal.
North star: % of unique landing-page visitors who complete the signup action, measured in your analytics tool over a consistent window (7 days is a common starting point — adjust to match your traffic volume).
North star: % of signups who reach the aha moment within a defined window (7 days is a common starting point — adjust based on your product's typical time-to-value).
North star: % of failed subscription payments that recover within a fixed window, measured from your billing provider's failed-payment event to a paid/captured payment event.
Concepts
| North star | The one real-world outcome that counts as progress — stable, measurable, and owned by goal.md. |
| Guardrail | A constraint that stops a cheap win from quietly degrading quality (e.g. "no engagement bait", "bounce rate must not rise"). |
| Lever | A mechanism you believe moves the north star — something you can pull, like hook quality or signup friction. |
| Lever portfolio | The archive of all levers in run.md, each tagged candidate, promising, proven, exhausted, or retired. |
| Action | One concrete step on a lever this run — a rep of a proven recipe, or a fix to something in the path. |
| Diagnostic metric | A measurement that explains why a lever did or did not move the north star. |
| Hypothesis | A falsifiable prediction attached to one lever, recorded in Superdense before the outcome is known. |
| Experiment | The Superdense record binding a hypothesis to one or more runs/artifacts and a reward window. |
| Mode | Each run is either explore (test an uncertain hypothesis) or exploit (rep a supported one on a proven lever). |
| Gate | The reusable completion contract that says what must be true before a run is called complete. |
Setup steps
- Get Superdense — install the CLI with `npm i -g @nimrobo/superdense`. It is the durable store for your sessions, artifacts, and reward snapshots.
- Make a folder for your outcome. One folder manages one outcome loop.
- Fire up your agent in that folder — Codex or Claude Code.
- Tell it which outcome pack to build. Pick one above and ask your agent to set it up.