beto-frega-skills

Plugin beto-frega-skills from marketplace beto-frega-skills in BetoFrega/skills. Commands read from the repo's manifest — today.

Skills (54)

• advisory

  Consult for stalls, repeated errors, complex algorithms, unresolved cross-layer interactions, or consequential decisions, diagnoses, plans, and reviews.

• autonomous-decisions

  Resolve a specified set of decisions through independent deliberation and evidence-based judgment.

• beto-mode

  Read first for multi-step, design, implementation, debugging, planning, or review work. Routes the task to working principles and workflow skills.

• check-feasibility

  Check real capabilities before recommending a solution.

• code-review

  Review branch, PR, or in-progress changes from a commit, branch, tag, or merge-base across standards, spec, performance, observability, and accessibility.

• codebase-design

  Shared vocabulary for designing deep modules. Use when the user wants to design or improve a module's interface, find deepening opportunities, decide where a seam goes, make code more testable or AI-navigable, or when another skill needs the deep-module vocabulary.

• consolidate

  Consolidate canonical product knowledge, ideas, behavior, decisions, increments, and evidence; route technical decisions and reconcile delivery, exposure, and flags.

• correct

  Sweep a repository's history for mistakes agents keep repeating and make each one impossible to repeat. Use for /correct as a periodic audit.

• decide

  Present one pending decision with a recommendation.

• demo-day

  Prepare a feature-review demo for a demo day — a story planned with the user, demo script, setup scenarios that seed the environment and print test data and users, a rehearsal recording each feature's demonstrability, and polished demo videos.

• diagnosing-bugs

  Diagnose hard bugs and performance regressions. Use when debugging failures or slowness.

• domain-modeling

  Build and sharpen domain terms and relationships, or write/edit the domain glossary in CONTEXT.md. Reading existing vocabulary alone does not require this skill.

• esds-compress

  Compress input using Exponential Semantic Density Scaling (ESDS). Use for ESDS compression or dense semantic payloads of state changes, architectural decisions, outcomes, and blockers.

• esds-decompress

  Decompress ESDS payloads into compact human-readable text. Use to expand, explain, or understand a payload without its notation.

• explain-pr

  Explain large or complex pull requests as a navigable HTML guide, tracing user journeys from entrypoints through changed code. Use when asked to understand a PR or build a guided walkthrough of its changes.

• grill-me

  An interview, one question at a time, to sharpen a plan or design.

• grill-with-docs

  Read product and domain context, grill selected increments, and preserve confirmed knowledge for consolidation and delivery planning.

• grilling

  Clarify and stress-test a selected plan, decision, idea, or product increment through scoped interview rounds. Use for grilling or consequential uncertainty.

• handoff

  Prepare a concise continuation brief for the current task.

• host-network

  Request host execution when sandbox network restrictions cause connection failures or misleading CLI authentication errors, including gh.

• implement-spec

  Implement an entire specification and its ticket graph as one PR.

• implement

  Implement selected work from a spec or tickets.

• investigate-low-density

  Investigate flagged or suspected low information density in logs, documentation, instructions, or responses. Use for deep analysis of repetition, noisy output, unselected dumps, or reducible content, including issues labeled `low-info-density`; this skill finds causes and safe reductions but does no

• last-responsible-moment

  Apply Last Responsible Moment to ongoing work to curb overplanning, premature commitments, or risky deferrals.

• migration-planning

  Plan and execute infrastructure, architecture, or database migrations.

• model-selection

  Select models and effort for implementation, review, recommendations, or delegation using capability, cached costs, and unresolved reasoning.

• next-steps

  Identify remaining completion requirements, meaningful learning to review, and ready successor work.

• permission-preflight

  Forecast or batch task authorizations as numbered requests for selective approval.

• plan-increment

  Plan selected product increments as vertical slices and executable work in the configured tracker, including bounded discovery, dependencies, acceptance, and exposure strategy.

• pr

  Use when writing a PR body.

• proceed

  Continue executing the agreed plan for the current task.

• prototype

  Prototype a state model, business logic, or UI with throwaway code that answers a design question.

• prune-agent-docs

  Propose independently reviewed reductions to agent documents, then apply user-authorized cuts.

• recommend

  Compare viable alternatives and recommend a choice; resolve missing information that could change it.

• record-decision

  Record an approved decision in its canonical project or task record.

• report-low-density

  Prepare one delegated low-information-density flag as a reproducible tracker issue; publish only with the user's explicit request or approval. Use neither for investigation nor remediation.

• report-skill-lesson

  Prepare an issue on an installed plugin's repository when one of its skills misled the agent or needs a change; publish only with the user's approval. Use instead of editing plugin skills in place.

• research

  Investigate a question against high-trust primary sources and capture findings at the configured research location. Use when the user wants a topic researched, docs or API facts gathered, or reading legwork delegated to a background agent.

• resolving-merge-conflicts

  Use when you need to resolve an in-progress git merge/rebase conflict.

• review-learnings

  Review evidence and learning across product work, resolve consequential choices through grilling, and feed supported knowledge and work implications into canonical records.

• review-triage

  Triage code-review findings, determine PR approval versus feature activation readiness, and require verified tickets for deferred work. Use to apply or explain this process in an existing review.

• scope-check

  Compare current work with the agreed task scope.

• setup-beto-frega-skills

  Configure global rules and project skill conventions, inside or outside the repository. Run explicitly before first use or for configuration gaps.

• ship-main

  Deliver current-task changes to main with validation, automatic reviews, and finding triage.

• simplify-plan

  Simplify an implementation or delivery plan using YAGNI and Last Responsible Moment while preserving proportional SOLID design. Use when a plan has speculative abstractions, unnecessary layers, excessive phases, or commitments beyond the next useful increment.

• status

  Summarize the current task's progress and blockers.

• tdd

  Test-first features, bug fixes, integration tests, or red-green-refactor requests at agreed seams.

• triage

  Move issues and external PRs through a state machine of triage roles, categorise, verify, grill if needed, and write agent-ready briefs.

• verb-tense

  Choose verb tense and mood by when each event happens and how certain it is. Use when writing or reviewing any text, and when a reader asks for a tense review.

• verify

  Check whether the current task's claimed outcome is confirmed.

• wizard

  Wizard scripts for human-only provisioning, credential/CI setup, unfamiliar dashboards, migrations, or cutovers; use only for steps the agent cannot perform.

• write-adr

  Write, revise, or supersede architecture decision records (ADRs) documenting an architectural choice and its rationale.

• writing-for-agents

  Write agent-facing documents. Use when creating or editing skills, AGENTS.md, or CLAUDE.md.

• writing-for-humans

  Load before writing any reader-facing text, including messages, and follow its rules as you write. Also use to rewrite or audit existing prose for plain language, clarity, readability, or actionability.

Subagents (1)

• beto-agent

  Delegate for implementation and helper assignments handed off during beto-mode work. Reads beto-mode and the principles the assignment triggers before acting. Use a fresh agent per assignment.

Hooks (2)

• SessionStart

  bash "${CLAUDE_PLUGIN_ROOT}/hooks/session-start.sh"

• UserPromptSubmit

  bash "${CLAUDE_PLUGIN_ROOT}/hooks/prompt-reminder.sh"

Similar plugins