skdd-commons

Interrogate a product for its core value — name the observable aha moment, measure time-to-aha, or honestly conclude there is none yet. Use when starting Growth Circle work on any repo, when someone asks "how do we grow this", when GROWTH.md is missing or its Aha section is empty, or before any activation/retention/distribution work is proposed. This is the gatekeeper skill — every other growth skill depends on its output.

Compile a product's earned, evidence-backed claims from GROWTH.md into a public VALUE.md — the proof-of-value surface that agent-mediated discovery reads — with attestation levels, expiry dates, and honest retractions. Use when creating or refreshing a VALUE.md, when marketing copy needs grounding in evidence, when preparing a product to be evaluated by assistants and their principals' agents, or when a published claim has gone stale.

Run the growth experiment cadence — propose ranked by keystone impact, pre-register before launch, evaluate by the committed rule, keep/kill without relitigating, log everything. Enforces the virality gate (get-users-to-get-users proposals are parked until activation and retention run green) and the alienation test (auto-kill anything a fully-informed user would resent). Use when the growth cadence comes due (weekly/biweekly), when someone proposes "a growth hack", or when experiment results need an honest verdict.

Use SkDD to forge project-specialized adaptations of the Growth Circle skills — local skills that carry this product's keystone, event names, query paths, contexts, and cadence baked in, so future sessions (and cheaper models) run the circle without rediscovering it. Use when the generic pack has run against a project a few times and keeps re-deriving the same context, when handing growth work to scheduled or lighter-weight agents, or when a project's circle has stabilized enough to be worth freezing as skills.

Make the growth circle self-running — schedule its recurring checks (loop cadence, keystone sunset, first-light freshness, AEO probes, trust-ledger review) using whatever scheduling primitive the current harness offers, and record every schedule in GROWTH.md so it survives harness changes. Use when the circle's gates are open but nothing recurs, when a keystone sunset or instrumentation check exists only as a date nobody watches, after run-the-loop establishes a cadence, or when growth work keeps happening "whenever someone remembers".

Turn growth-team values into working artifacts — role scorecards, interview kits with listen-fors — and map which rings of the growth circle run on agents versus humans, with explicit approval boundaries. Use when hiring a first growth person, when staffing growth work across a human/agent team, when a growth hire is failing and nobody can say why, or when deciding what an autonomous agent may and may not do to users.

Detect when a playbook that works in one context is failing in another — market, language, platform, segment, or agent-mediated channel — and run the 99/1 reset protocol with native-informed deviations. Use when the same keystone reads very differently across segments, when entering a new market or platform, when metrics are flat somewhere the product "should" work, or when deciding how the product shows up in assistant recommendations.

Derive a product's keystone metric — who × does-what × N times × window — by working backwards from engaged users, then stress-test it for causation and gameability before anyone reorganizes around it. Use when a team asks "what's our North Star", when activation work has no target, when an existing keystone was adopted from a blog post rather than cohort evidence, or when GROWTH.md has an Aha but an empty Keystone.

Wire minimal, consent-first instrumentation into a codebase — an event schema for the keystone path, cohort queries, and experiment scaffolding on whatever stack exists. Use when GROWTH.md has an Aha but no verified instrumentation, when find-the-keystone stalls for lack of data, when a team is flying on vibes, or when an analytics setup has grown into an unusable event swamp and needs cutting back to the path that matters.

Harvest a team's beliefs about why users behave, convert each into a testable claim, and run the cheapest disproving test — building an append-only lore ledger of confirmed/busted/undecidable verdicts. Use when roadmap arguments run on gut feeling, before any big bet built on an untested assumption, on a quarterly hygiene cadence, or when someone says "users churn because…" without a link attached.

Run an adversarial pre-mortem on a plan, spec, or architecture before anyone builds it — argue the case of the smartest person who opposes it, not a polite review. Use when a significant design is about to be implemented, before a refactor or migration, when someone shares a plan and asks "thoughts?", or when a decision would be expensive to reverse.

Closed-loop delegation — define done as an observable behavior, boot the real app, drive it like a user, fix what breaks, and verify by using it again before reporting. Use when implementing any feature or fix with a runtime surface, when tempted to report "it should work now", or when a task will span a long unattended stretch and the result must be trustworthy without supervision.

End-of-session artifact extraction — convert what a session produced into durable assets (a skill, a DESIGN.md, a checklist, a codebase map) and register them in the colony. Use when a substantive working session is ending, when someone says "save this workflow" or "remember how we did this", after a hard-won debugging victory, or whenever model access is scarce and the judgment spent should outlive the session.

Detect a plan-shaped request and surface the problem one altitude level above it before executing. Use when a request arrives as a prescriptive implementation step ("add caching to this endpoint", "add an index", "wrap it in a retry") with no symptom, metric, or goal attached, or when the obvious fix appears to treat a symptom rather than a cause.

Fan wide work out to parallel agents — one lane and one definition of done per agent — then adversarially verify the findings and synthesize a single answer. Use when work is wide rather than deep (full-site audits, migrations across many call sites, research from multiple angles, review across dimensions), when a linear walk would take hours, or when independent perspectives materially reduce the chance of a wrong conclusion.

Subtraction-first review of a page, copy deck, API, schema, or config — judge the real rendered artifact, inventory everything it contains, and rank what should be removed. Use when something reads as bloated, generic, or untrustworthy; before a launch; when asked to "make it better" or "polish this"; or when a design/doc/API has grown by accretion and nobody remembers why.