Skip to content
Adaptive Skills Docs
Esc
navigateopen⌘Jpreview
On this page

Workflow recipes

Combine a small number of Adaptive Skills for common delivery, debugging, design, and governance tasks.

Recipes are starting patterns, not automatic pipelines. Run one skill at a time, inspect its output, and decide whether the next skill is still necessary.

Deliver a new feature

  1. intent-clarification — confirm outcome, constraints, success, and unresolved ambiguity.
  2. feature-value-governance — check whether the feature deserves investment when value is uncertain.
  3. feature-planning — define the smallest useful slice and its proof.
  4. lean-implementation — implement only the approved slice.
  5. testing — choose and run risk-appropriate proof.
  6. handoff-summary — return verified state and the next safe step.

Stop early when: intent or value remains unresolved. Do not convert ambiguity into implementation scope.

Debug a production issue

  1. workflow — state the incident boundary and closure proof.
  2. debugging — reproduce, isolate, fix, and guard recurrence.
  3. testing — validate the failure mode and nearby risk.
  4. communication — explain impact, resolution, and remaining uncertainty.
  5. handoff-summary — preserve verified incident context.

Stop early when: the issue cannot be reproduced. Record the evidence gap instead of guessing at a fix.

Review a product experience

  1. ux-strategy — clarify the experience goal and plausible directions.
  2. heuristic-audit — identify usability failures in the current interface.
  3. ux-writing — improve labels, guidance, errors, and decision language.
  4. triad-check — reconcile product, design, and engineering consequences.

Optional: add ux-provocation before commitment when the team has converged too quickly on one direction.

Make an architecture decision

  1. domain-language-alignment — establish shared terms.
  2. architecture-review — compare boundaries, coupling, complexity, and maintenance cost.
  3. premortem — identify failure conditions and warning signals.
  4. triad-check — review cross-functional consequences when the decision affects the product experience.
  5. communication — record the decision and trade-offs.

Govern a feature portfolio decision

  1. revenue-lever-mapping — make the intended value mechanism explicit.
  2. opportunity-tree-alignment — connect the bet to an outcome and opportunity.
  3. feature-complexity-audit — estimate permanent carry cost.
  4. feature-value-governance — make the auditable investment judgment.
  5. sunset-decision — use only for an existing feature under keep-or-remove review.

Use governed knowledge safely

  1. knowledge-source-evaluation — assess whether the source is suitable and at what maturity.
  2. restricted-context-check — review leakage, injection, permission, and contamination risks.
  3. knowledge-conflict-resolution — resolve decision-relevant disagreement between accepted sources.

Boundary: these skills do not grant access, override permissions, or turn untrusted content into authoritative instructions.

Close a long work session

  1. checkpoint-review — decide whether to continue, stop, or hand off.
  2. qa-review — check consistency and operational risk if meaningful work changed.
  3. handoff-summary — package verified state without dragging the full session forward.

Next steps

  • Use Choose the right skill to adapt a recipe to the current bottleneck.
  • Read How to use a skill before invoking the first skill.
  • In AletheIA, keep the Work Slice as the governing frame and use skills only inside that boundary.

Was this page helpful?