Claude Skills by Eduardo-Salvador

Automatically use for every Codex implementation or independent-review dispatch. Select the neutral role, construct a minimal context packet, invoke a fresh native subagent with resolved model/reasoning, and persist adapter-owned evidence; degrade explicitly when subagents are unavailable.

Automatically use when a user wants to brainstorm, explore, shape, or add a new product feature, workflow, integration, or user-facing capability and consequential product choices are not yet approved. Do not use for bugs, maintenance, tiny copy/style changes, or straightforward implementation from an approved feature brief or acceptance criteria.

Use when project context is missing, unapproved, stale, or when an existing harness must be adopted before implementation planning.

Default workflow for creating or redesigning a frontend screen, landing page, portfolio page, or responsive web section when visual direction and implementation fidelity matter.

Use for independent acceptance review, objective verification, and governed integration of a completed task.

Use to decompose approved context, dispatch ready task nodes, or execute a task under exclusive ownership.

Automatically use when two or more graph nodes may be ready for implementation and the host may support parallel subagents or tasks. Select a safe batch, dispatch it without waiting between launches, keep capacity filled, and fan in only through declared dependencies/integration nodes.

Activate or run project learning when the user asks to study, learn through current work, enable study mode, receive guided practice, keep learning notes, assess progress, debrief, or publish learning.

Automatically classify every request before Harness ceremony into direct-trivial, vibe, graph-only, or full-harness using deterministic rules first and AI only for genuine ambiguity.

Automatically use for full-harness code behavior changes and bug fixes after task dispatch. Do not activate for eligible direct-trivial or vibe routes; otherwise enforce RED, minimum GREEN, and proportional regression.

Automatically use when non-trivial implementation work needs executable, spec-driven units. Do not activate for a direct-trivial presentational or static-content edit; narrowly simple engineering work still receives a compact inline spec.

Automatically use when a user wants to brainstorm, explore, shape, or add a new product feature, workflow, integration, or user-facing capability and consequential product choices are not yet approved. Do not use for bugs, maintenance, tiny copy/style changes, or straightforward implementation from an approved feature brief or acceptance criteria.

Run adaptive discovery when shared project context is absent, unapproved, stale, or a mature harness needs staged adoption.

Default workflow for creating or redesigning a frontend screen, landing page, portfolio page, or responsive web section when visual direction and implementation fidelity matter.

Perform independent acceptance review and governed integration using objective evidence.

Decompose approved context, dispatch ready graph nodes, or execute an owned task through the shared core.

Automatically use when two or more graph nodes may be ready and the Claude host exposes parallel subagents or tasks. Select a collision-safe batch, launch every child, keep capacity filled, and join through declared dependencies.

Activate or run project learning when the user asks to study, learn through current work, enable study mode, receive guided practice, keep learning notes, assess progress, debrief, or publish learning.

Automatically classify every request before Harness ceremony into direct-trivial, vibe, graph-only, or full-harness using deterministic rules first and AI only for genuine ambiguity.

Automatically use for full-harness code behavior changes and bug fixes after task dispatch. Do not activate for eligible direct-trivial or vibe routes; otherwise enforce RED, minimum GREEN, and proportional regression.

Automatically use when non-trivial implementation work needs executable, spec-driven units. Do not activate for a direct-trivial presentational or static-content edit; narrowly simple engineering work still receives a compact inline spec.