brokenrobot-xyz
- 40 skills
- 0 followers
- 15 hours ago last updated
- ▌ Formatting Changelogs 2 · brokenrobot-xyz bundleFormat the project's CHANGELOG.md before a release: group the unreleased entries by change type, order the groups, and stamp the release heading. Use this skill when preparing a changelog for release.
- ▌ Prompt Quality Criteria · brokenrobot-xyz bundleSupplies criteria groups B–G for grading a Markdown prompt that steers Claude — model-specific prompting, general prompting, hallucination guards, output consistency, injection and jailbreak defenses, and prompt-leak defenses. Returns the criteria; the caller scores against them. Use when reviewing a skill, a subagent definition, or another agent-facing prompt, or when the user asks what to check in one.
- ▌ Committing Conventionally · brokenrobot-xyz bundleStages working-tree changes and authors one Conventional-Commits commit, resolving the commit vocabulary from the host project's .brokenrobot-xyz/commits.json when present and from built-in defaults when not. Use whenever the user asks to commit work.
- ▌ Writing Simplified Technical English · brokenrobot-xyz bundleRevises agent-facing prose — SKILL.md bodies, agent definitions, spec and planning artifacts, and technical documentation — so that an agent cannot read a sentence two ways. Applies twelve conventions adapted from ASD-STE100 Simplified Technical English, Issue 9: it takes the rules that remove ambiguity, and it omits the standard's controlled dictionary and its sentence-length caps, so it is an adaptation rather than a conforming implementation. Also checks prose and reports violations without editing it. Use when authoring or tightening a skill, an agent definition, a proposal, a design, a task list, or a technical document, and when asked to make instructions clearer, less ambiguous, or more precise. Does not govern published marketing copy, commit messages, or code.
- ▌ Updating Dependencies · brokenrobot-xyz bundleUpdates, upgrades, and bumps npm dependencies in the host project — detects which packages are outdated, buckets them into patch/minor/major, researches the bumps the user selects, and applies only the bumps the user approves. Use when refreshing, updating, or upgrading dependencies, when packages are behind latest, or when npm outdated needs acting on. Edits package.json, the lockfile, and the source files an approved migration requires, but never commits.
- ▌ Improving Claude Skills · brokenrobot-xyz bundleAutonomously improves a Claude Code skill in a review→fix→re-review loop: each round invokes the reviewing-claude-skills review non-interactively, applies every blocking finding — High and structural findings included — without per-fix approval, commits the round, and re-reviews, until the review's verdict is acceptable, the blocking findings plateau, or the round cap is hit. Advisory findings are reported, never chased. Asks one question only: confirming the intent brief at kickoff. Use when the user asks to improve a skill autonomously, in a loop, or until it passes review.
- ▌ Reviewing Claude Skills · brokenrobot-xyz bundleReviews a Claude Code skill — its SKILL.md, evals, and referenced files — in two passes. Structure first: a workflow whose shape fails High stops at a structural verdict with a redesign recommendation. Otherwise a full detail sweep against skill-authoring and prompting best practices plus the host project's conventions produces a severity-ranked gap analysis ending in a computed verdict — acceptable, or not yet — honoring the skill's recorded waivers, optionally applying approved fixes. Use when the user asks to review, audit, or improve a skill.
- ▌ Improving Claude Subagents · brokenrobot-xyz bundleAutonomously improves a Claude Code subagent definition in a review→fix→re-review loop: each round invokes the reviewing-claude-subagents review non-interactively, applies every blocking finding — High and structural findings included — without per-fix approval, commits the round, and re-reviews, until the review's verdict is acceptable, the blocking findings plateau, or the round cap is hit. A gated verdict recommending a different artifact form — 'this should be a skill' — ends the loop for the human instead of being applied. Advisory findings are reported, never chased. Asks one question only: confirming the intent brief at kickoff. Use when the user asks to improve a subagent autonomously, in a loop, or until it passes review.
- ▌ Reviewing Claude Subagents · brokenrobot-xyz bundleReviews a Claude Code subagent definition — its frontmatter, body, declared tools, and the siblings it competes with for routing — in two passes. Structure first: a definition whose form or shape fails High stops at a structural verdict with a redesign recommendation. Otherwise a full detail sweep against subagent-authoring and prompting best practices plus the host project's conventions produces a severity-ranked gap analysis ending in a computed verdict — acceptable, or not yet — honoring the subagent's recorded waivers, optionally applying approved fixes. Use when the user asks to review, audit, or improve a subagent or an agent definition.
- ▌
- ▌ Preparing Release Branches · brokenrobot-xyzCreates the release branch for a milestone and verifies that the branch builds. Use when cutting a release.
- ▌ Refreshing Visual Baselines · brokenrobot-xyzRefreshes the visual-regression baselines after an intentional visual change. Use when the snapshot suite reports an intended difference.
- ▌ Generating Alt Text · brokenrobot-xyzGenerates alt text for images referenced in Markdown documentation.
- ▌ Syncing Shared Branches · brokenrobot-xyzSynchronizes a shared feature branch with its remote and resolves divergence. Use when a shared branch falls behind the trunk.
- ▌ Sorting Imports · brokenrobot-xyz bundleSorts the import block of JavaScript and TypeScript files into groups — runtime built-ins, third-party packages, local paths — with each group alphabetized. Use when imports fail a lint rule or read unordered.
- ▌ Foo Baz · brokenrobot-xyzRenames image assets to content-hashed filenames and updates each reference to them. Use when cache-busting static assets.
- ▌ Escaping YAML · brokenrobot-xyz bundleEscapes user-supplied strings for safe embedding in YAML documents: quotes, special characters, multi-line blocks, and type-coercion traps. Use when the user asks to escape, quote, or safely embed a value in YAML.
- ▌ Auditing CSS Tokens · brokenrobot-xyzAudits a stylesheet folder for hard-coded values that duplicate an existing design token — colors, spacing, radii — and reports each with the token to use instead. Use before a release or after importing third-party CSS.
- ▌ Triaging Bug Reports · brokenrobot-xyzTriages an inbound bug report — assigns severity, labels the affected area, and drafts a first response asking for whatever reproduction detail is missing. Use when a new bug report arrives.
- ▌ Archiving Notes · brokenrobot-xyz bundleArchives a project's stale meeting notes fully autonomously — moves every note older than the cutoff into the archive folder without asking any questions. Use when the user asks to archive, clean up, or tidy old notes.
- ▌ Tagging Releases · brokenrobot-xyz bundleTags a release in git following the host project's release conventions: derives the tag name from the version, writes the annotation from the changelog, and pushes the tag. Use when the user asks to tag or cut a release.
- ▌ Staging Release Notes · brokenrobot-xyzStage the release notes and update the changelog header before you tag the release. Use before tagging a release.
- ▌ Generating Sitemaps · brokenrobot-xyz bundleGenerates a sitemap.xml for a static site from its built HTML — collects page URLs, assigns priorities, and escapes entities per the bundled rules. Use after a build when the sitemap is missing or stale.
- ▌
- ▌ Formatting Changelogs · brokenrobot-xyzNormalizes CHANGELOG.md to Keep a Changelog form — one section per version, entries grouped under Added, Changed, and Fixed, newest first. Use when a changelog drifts from the format or before a release.
- ▌ Formatting Tables · brokenrobot-xyz bundleReformats Markdown tables in documentation files — normalizes column alignment and pads cells — following the bundled format guide. Use when a document's tables render unevenly or fail a format check.
- ▌ Counting Words · brokenrobot-xyzCounts words, sentences, and reading time for Markdown files, excluding code blocks and frontmatter. Use when checking an article against a length budget.
- ▌ Summarizing Changelogs · brokenrobot-xyz bundleSummarizes a package's CHANGELOG.md between two versions into a short upgrade note — breaking changes first, then features, then fixes. Use when deciding or documenting a dependency bump.
- ▌ Mapping Redirects · brokenrobot-xyzMaintains the redirect map for moved pages — adds entries for renamed or deleted URLs and flags chains and loops. Use when pages move or a 404 report comes in.
- ▌ Checking Alt Text · brokenrobot-xyzChecks every image in Markdown and HTML files for alt text — flags missing, empty, and filename-echo alt attributes with file and line. Use before publishing content or during an accessibility pass.
- ▌ Rotating API Keys · brokenrobot-xyzRotates a service's API key — generates the replacement, updates each consumer, and retires the old key after the overlap window. Use on a rotation schedule or after a suspected exposure.
- ▌ Checking License Headers · brokenrobot-xyzChecks source files for the required license header — reports files where the header is missing, stale, or malformed, with the exact first divergent line. Use during an audit or before a release.
- ▌ Syncing Translations · brokenrobot-xyzSynchronizes the host project's locale files against the source-of-truth English catalog — buckets entries into identical, fuzzy, and missing, auto-applies the configured categories, and gates the rest on user approval. Use when locale files drift behind the English catalog or translations need syncing.
- ▌ Redacting Logs · brokenrobot-xyzRedacts secrets and personal data from log files before they are shared: API keys, tokens, email addresses, and IP addresses, replacing each with a typed placeholder. Use when the user asks to redact, sanitize, or scrub a log.
- ▌ Summarizing Minutes · brokenrobot-xyz bundleSummarizes a meeting's minutes into a short brief: decisions, owners, deadlines, and open questions, in the format the style guide defines. Use when the user asks to summarize minutes or produce a meeting brief.
- ▌ Linting Frontmatter · brokenrobot-xyz bundleChecks Markdown frontmatter blocks against a required-key list — reports missing keys, empty values, and unknown keys with file and line. Use when frontmatter errors break a site build or a content check fails.
- ▌ Pruning Feature Flags · brokenrobot-xyzRemoves fully-rolled-out feature flags from a codebase — deletes the flag definition, inlines the enabled branch, and cleans the config. Use when a flag has sat at full rollout for a release cycle.
- ▌ Publishing Docs · brokenrobot-xyzBuilds the documentation site and publishes the output to the gh-pages branch. Use when merged docs changes need to go live.
- ▌ Validating Feeds · brokenrobot-xyzValidates RSS and Atom feeds against their specifications — well-formedness, required elements, and date formats — and reports each violation with its element path. Use when a feed reader rejects a feed or after changing feed generation.
- ▌ Archiving Stale Branches · brokenrobot-xyzArchives stale git branches after checking their merge status and open pull requests.