Claude Skills by yeaight7

Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for merge, PR, or cleanup

Create and edit JSON Canvas files (.canvas) with nodes, edges, groups, and connections. Use when working with .canvas files, creating visual canvases, mind maps, flowcharts, or when the user mentions Canvas files in Obsidian.

Use when triaging pull requests, stale branches, or review bottlenecks and you need the most relevant actions and merge risks from the freshest repository state available.

Relay questions and tasks to a local Claude CLI using one-shot subprocesses per ask.

Relay questions and tasks to a local Codex CLI using one-shot subprocesses per ask.

Relay questions and tasks to a persistent local Gemini ACP session with cross-turn context.

Check version bump, changelog, compare range, package metadata, validation commands, publish steps, and rollback plan before cutting a release.

Use when starting feature work that needs isolation from current workspace or before executing implementation plans - creates isolated git worktrees with smart directory selection and safety verification

Inspect and test web UIs using screenshots, browser automation, accessibility snapshots, and console/network signals with structured visual evidence collection.

Use when creating new skills, editing existing skills, or verifying skills work before deployment

Use when starting work in an unfamiliar area of a codebase, spawning a subagent that needs targeted file context, a first search pass missed the relevant file, or the file scope of a task is unclear.

Use when about to refactor, rename, or remove code, assessing dead code before deletion, or unsure of the blast radius of a change.

Use when designing tool definitions for a new agent or subagent, an agent shows high retry rates, ambiguous tool invocations, or silent failures, or an existing agent harness needs a quality review.

Use when starting work in an unfamiliar area of a codebase, spawning a subagent that needs targeted file context, a first search pass missed the relevant file, or the file scope of a task is unclear.

Use when about to write a new doc, README section, or ADR, updating existing documentation with unclear scope, or spawning a documentation subagent that needs file context.

Use when ending a session, handing a task back to the user, or approaching the context limit with work unfinished -- the next session or engineer must be able to resume the work instantly.

Use when about to add a new helper, utility, or abstraction, a task sounds like a solved problem, a new external dependency is being considered, or custom code is proposed without checking what already exists.

Use when starting work in a repository with Agent Powerups installed, when a task may match a reusable local skill, command, workflow, hook recipe, AGENTS.md template, or MCP feature.

Address actionable GitHub pull request review feedback. Use when the user wants to inspect unresolved review threads, requested changes, or inline review comments on a PR, then implement selected fixes.

Use when evaluating prompts, LLM outputs, red-team suites, or model behavior with local eval configs and safe provider/cost controls.

Use when creating or reviewing red-team eval plugins, attack templates, grader rubrics, safety fixtures, or model-risk test metadata.

Use when designing, running, debugging, or hardening deterministic eval suites for agent skills, prompts, tool workflows, or MCP-backed cases.

Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always

Use when a hook is about to be enabled or modified -- a hook recipe proposed for activation, a pre/post-tool or lifecycle hook added to agent settings, a git pre-commit hook added to a repo, or a third-party hook snippet pasted in from outside.

Use when an MCP server is about to be enabled, added to agent config, or upgraded -- a staged server config is a candidate for activation, a third-party server is proposed from a registry, or an existing entry changes transport, credentials, or tool surface.

Use when creating or reviewing red-team eval plugins, attack templates, grader rubrics, safety fixtures, or model-risk test metadata.

Use when about to commit, push, or publish -- staged changes touch config or environment files, generated artifacts (relay sessions, logs, build output) are being added, or the session handled credentials even indirectly.

Use when designing or using MCP-backed structured code search with search, AST query, symbol inventory, and bounded extraction workflows.

Use when modifying agent config files, committing configuration changes, onboarding a repository with existing agent configs, or running a periodic config hygiene check.

Use when designing tool definitions for a new agent or subagent, an agent shows high retry rates, ambiguous tool invocations, or silent failures, or an existing agent harness needs a quality review.

Use when an agent has modified logic, API routes, or data-transformation code, a bug was just fixed and must not be reintroduced, or behavior must stay identical across execution paths such as sandbox versus production.

Use when a working code path feels bloated, noisy, or over-abstracted, when asked to clean up or deslop AI-generated code, or when a reviewer-only anti-slop pass is requested.

Use when API documentation may have drifted from the implementation -- endpoints changed without a spec update, integration partners report mismatches, or a PR touches routes, params, or response shapes.

Use when a significant architectural choice is being finalized, revisited, or reversed -- technology selection, structural patterns, or trade-offs that future maintainers or agents might unknowingly undo.

Use when a codebase carries over-engineered abstractions, unnecessary layers, or redundant logic that should be collapsed without changing behavior.

Use when the user wants end-to-end autonomous execution from a brief idea to working, verified code -- multi-phase work spanning requirements, planning, implementation, QA, and validation. Not for brainstorming, single focused changes, or quick fixes.

Use when a new or more complex ML model is proposed and its value over simple baselines is not yet demonstrated -- before approving a new architecture or replacing an existing heuristic.

Use when reviewing BigQuery spend, query failure patterns, or scan inefficiencies -- identifying which jobs, users, or projects drive cost, or preparing optimization recommendations for a cost review.

Use when starting any creative work — creating features, building components, adding functionality, or modifying behavior — before any implementation begins.

Use when routing a prompt to a local provider CLI for a second opinion, review, or plan -- you are about to call a provider directly, need the response saved for later, or want a single consistent way to invoke codex/gemini/claude.

Use before submitting a PR or considering a task done to evaluate the blast radius of your changes -- you touched a public signature, a shared utility, a schema, or added config and need to know what else is affected.

Use when a CI pipeline has failed and the real error must be extracted from thousands of lines of noisy logs.

Use when a wide refactor or migration is too large to ship as one change — API renames across many files, framework or config migrations, or mechanical codemod-backed refactors that need reviewable batches.

Use when a subsystem needs persistent context for future agents or engineers -- its boundaries keep being rediscovered the hard way, invariants keep getting violated, or sessions repeatedly ask the same questions about it.

Use when a task runs long and the context window is filling -- large files are being read whole, command output is flooding the transcript, or responses are getting verbose and slow.

Use when starting work in an unfamiliar area of a codebase, spawning a subagent that needs targeted file context, a first search pass missed the relevant file, or the file scope of a task is unclear.

Use when adding or reviewing data quality tests for dbt models in warehouse-backed analytics projects. Covers dbt generic tests, singular tests (assert_*.sql), accepted-values macros, dbt_utils patterns, cross-system consistency tests, and warehouse-oriented validation. Use when writing data tests, creating assert_*.sql files, testing business logic, or validating referential integrity.

Use when reviewing how data is split into train/validation/test sets -- especially with time-series data, repeated entities (users, patients, sessions), or imbalanced targets.

Use when reviewing a new or changed dbt incremental model -- the strategy looks over-engineered, rebuild or delete behavior is unclear, or the model deviates from how the rest of the repo handles incrementals.

Use when a dbt change needs preflight before a PR, review, or merge -- changed models, snapshots, seeds, macros, or semantic YAML whose blast radius and test coverage are not yet known.