docs

Rewrite technical product documentation to strip AI slop and conform to the CTRT topic-type model (Concept, Task, Reference, Troubleshooting) plus Tutorial. Use whenever the user wants to deslop, tighten, edit, or rewrite technical docs — concept pages, task / how-to pages, reference material (API, CLI, config tables), troubleshooting pages, runbooks, tutorials, release notes, or product README sections. Trigger on phrases like "deslop", "make this doc less AI-sounding", "edit this docs page", "tighten my draft", or "rewrite this in CTRT", whether the user pastes prose or points at a file (.md, .mdx, .rst, .adoc, .txt, .docx). Scope is product/technical documentation only — do not use for marketing copy, blog posts, social posts, or general prose. Default behavior is to classify the input by topic type, then apply a heavy rewrite per the type-specific rules. Supports an optional passive mode (light-touch: word- and sentence-level fixes only, no restructuring — trigger on "passive", "light-touch", "words only"

Apply Seqera docs house structural conventions to a documentation page or selection — page introductions, prerequisites, troubleshooting placement, admonitions, and other structural elements. Use when the user asks to format, clean up, standardize, restructure, or fix the structure of a docs page, or when writing a new guide/tutorial that needs to follow house format. Each structural area has its own reference file under references/.

Generate OpenAPI overlay files for Seqera Platform API documentation. Use when working with API documentation updates, analyzing comparison overlays from Speakeasy, creating operations/parameters/schemas overlay files, or updating API documentation for new Platform versions. Triggers include requests to generate overlays, document new API endpoints, analyze API changes, or work with Seqera Platform OpenAPI specifications.

Format and style-check open source Nextflow and Wave changelogs to ensure consistency in voice, tense, attribution, linking, and component-based organization. Use when formatting a changelog in changelog/nextflow/ or changelog/wave/.

Format and style-check Seqera Platform changelogs (Cloud and Enterprise only) for consistency in voice, tense, component-based organization, and formatting conventions. Use when formatting a Platform changelog, checking changelog style for Cloud or Enterprise releases, or working with files in changelog/seqera-cloud/ or changelog/seqera-enterprise/. Does not apply to Wave, Nextflow, or Fusion changelogs — use changelog-os-formatter for those.

Turn feature sources (Jira epics, PRDs, design docs, GitHub PRs) into documentation updates for Seqera Platform docs. Use this skill whenever the user wants to: write docs for a new feature, update existing docs based on a Jira epic or ticket, draft doc changes from a PRD or design doc, review PRs and write matching documentation, do a gap analysis between current docs and shipped features, create a change summary or changelog for doc reviewers, or says anything like "I need to do doc updates for this feature" or "can you draft docs from these sources". Also use when the user provides a mix of Jira links, Google Doc links, GitHub PR links, or uploaded PDFs and wants documentation produced from them.

Triage the doc impact of a single Seqera release against three surfaces: environment variables (does the change add/rename/remove/alter a var listed in seqeralabs/docs under platform-enterprise_docs/enterprise/configuration?), visual assets (do UI screenshots, recorded videos, or brand marks in the docs go stale?), and internal doc links (are page renames, moves, removals, or anchor changes breaking cross-page links?). Always use when the user gives a PR URL, Jira key, or changelog entry and asks what it breaks in the docs, or says: "check impact of this release", "run an impact check", "does this PR break docs", "is this a rebrand", "what docs does this affect", "what needs updating after this change", or "impact assessment". Also trigger on descriptions of a UI refresh, rebrand, env var change, page rename, or page removal. Prefer this over docs-state-assessment (broad audit) and docs-feature (drafts new content) for single-release triage.

Bump hard-coded version numbers across the Seqera docs in the two places they recur: (1) enterprise point releases — six files in platform-enterprise_versioned_docs/version-XX.Y/ that pin Docker image tags and Kubernetes manifests to a released version, plus the cloud-side functionality matrix that mirrors the enterprise compatibility table, and (2) the Connect client used by Studios — Dockerfile ARG defaults, docker build commands, prerequisite "or later" lines, and container-image tag samples scattered across platform-cloud/, platform-enterprise_docs/, and the versioned snapshots. Use when the user says: "bump the connect version", "version bump for 25.3.5", "what files do I need to update for the enterprise release", "where is the connect client mentioned", "scan for version numbers", or invokes this skill directly.

Run editorial review on documentation files using specialized agents (voice-tone, terminology, punctuation, clarity). Use when you need to review documentation for style, consistency, tone, or formatting issues. Triggers include requests to review docs, check editorial quality, or run style checks on markdown files.

Assess the documentation state of a Seqera product area by cross-referencing GitHub PRs, Jira backlog, published docs on docs.seqera.io, and internal repo documentation. Produces a gap analysis showing what's shipped but undocumented, what's in flight and will need docs, what's planned, and what published docs may already be stale. Use this skill whenever the user asks to: assess documentation gaps, audit doc coverage for a project, compare what's shipped vs. what's documented, produce a "state of the docs" report, do a doc gap analysis, check what PRs need docs, figure out what's missing from the published docs, or says anything like "what docs do we need" or "how do the docs stack up". Also trigger when the user provides a GitHub repo URL and a Jira project and wants to understand documentation coverage or priorities. This skill focuses on the state of the work — what exists, what's in flight, what's missing — not on personas, information architecture, or where content should live in the doc site.