Claude Skills by tony

Rebase current branch onto trunk (origin/master or origin/main), predict and resolve conflicts

Use when discovering and confirming the next version from a project's own release scheme before cutting the release.

Use when cutting a release at an explicit version, including version files, lockfiles, changelog dates, and commit.

Use when rolling a newly published package release through every maintained consumer repository and verifying CI.

Use when pulling upstream source or cloning a dependency at its installed version to study internals, read code, or create pinned worktrees.

Use when screened review findings should be fixed on the current branch, usually as one verified commit per finding.

Use when screening or replying to pull-request review comments and bot findings; decide if each comment is real, worth fixing, deferred, or declined.

Use when the review-and-CI loop should continue through bounded rounds until no new actionable feedback remains.

Use when upgrading Ruff in one repository or a fleet and evaluating newly applicable rules before committing each change.

Take reading notes on a prose corpus — quotations with stable locators and margin notes, as a term source for extract

Answer a question from a finished study, citing back into its evidence, and say plainly when the study does not cover it

Set two term tables against each other — same concept different term, same term different concept, and what each covers that the other lacks

Adversarially judge a distilled ontology one candidate at a time — synonymy, polysemy, orphans, circular evidence, and coverage

Build the three ontology layers from a term table — types, members with discriminators, and the controlled vocabulary

Harvest terms and their definitions from a gathered corpus into terms.jsonl, each with the citation that supports it

Resolve and pin a corpus, inventory it, and record what was read and what was deliberately skipped, into sources.jsonl

Use when distilling an ontology, controlled vocabulary, or glossary of what a codebase, a repository set, or a written corpus calls things.

Write a study's brief and evidence documents, then enforce the brief's citation prohibition with the analyzer

Record a correction to a standing claim — what was believed, what it rested on, and why it failed — rather than editing it away

Re-check a finished study against the corpus it cites — refs that moved, files that vanished, quotations that no longer match, numbers that changed

Use when applying accepted self-improvement sweep findings to skills as individually gated edits and commits.

Use when mining prompt history to rank skill usage, count invocations, find never-invoked skills, sweep the corpus, or improve repeated instructions.

Use when confusion is only “huh,” “wait, what,” “sorry, you lost me,” or “slow down; explain simply and keep it short.”

Use when a branch may have drifted from its ticket or pull-request goal and needs its work and gaps reclassified.

Use when you came back to a branch/PR: the work you were in the middle of, where it left off, repo state, and what reviewers ask or the PR waits on.

Use when the user asks what is going on and needs a five-line summary of the session, branch, pull request, and ticket.

Gain situational awareness — read the branch, its diff, its PR, its tickets, and the project's own conventions, and report where the work stands

Use when writing, auditing, or trimming a SKILL.md, especially for bloated bodies, weak descriptions, or plugin-wide sweeps.

Use when scanning tracked repository files for AI slop, verbosity, brittle references, or low-value contributions.

Use when building two to four competing strategies in isolated git worktrees to compare them without landing commits.

Use when spiking a feature or fix in the working tree to prove it quickly before committing to an implementation.

Use when asked to spike, bake off, then spike again; rinse and repeat, keep spiking, converge on a design, and stop when nothing new fights back.

Use when Tailwind UI spacing, alignment, gaps, margins, padding, or toolbar and navbar rhythm looks inconsistent.

TDD bug-fix workflow — reproduce a bug as a failing test, find root cause, fix, and verify

Use when fixing a bug through strict xfail reproduction and needing proof that the test fails for the intended reason.

Use when updating one Terraform or OpenTofu provider across modules and refreshing root lockfiles without losing platforms.

Use when updating the Terraform or OpenTofu CLI across module constraints, version-manager files, CI, and images.

Use when Terraform or OpenTofu provider constraints, CLI pins, or lockfiles are stale across repository root modules.

Use when refreshing Terraform or OpenTofu root lockfiles within existing constraints while preserving platform coverage.

Use when asked to file a Linear issue or regression, draft a Jira epic, open a GitLab issue, or create a Shortcut story or tracker document.

Use when rebuilding a bloated tracker item whose epic or Jira ticket hard-codes implementation, restates a spec, or cites counts and line numbers.

Weave architecture — compare independent adversarial workers, then synthesize the best project architecture

Weave question — ask independent adversarial workers in parallel, then synthesize the best answer

Use when asked to brainstorm then refine, generate ideas then improve, or explore alternatives and synthesize the best one across passes.

Use when a creative prompt, design question, or open-ended problem needs multiple independent adversarial ideas.

Weave execute — compare independent adversarial implementations, then synthesize their best parts

Fix weave review findings — validate, add test coverage, fix, and commit each as atomic changes

Weave planning — compare independent adversarial plans, then synthesize the best one

Weave prompt — compare independent adversarial implementations, then pick the best approach

Use when asked to refine a draft, iterate on code until cleaner, or polish a README through repeated critique and improvement.