[AzAdvertizer](https://www.azadvertizer.net) (by Julian Hayward) tracks Azure governance capabilities — Policies, Initiatives (policy sets), and RBAC Roles — and publishes them as **downloadable CSV exports**. There is **no API**. This skill turns those CSVs into deterministic offline lookups: f

`c7search` wraps the Context7 v2 documentation API in a single static binary. It exposes two operations — resolve a library name to an ID, then fetch docs for that ID — with on-disk caching, secret redaction, retries, and semantic exit codes. Payload goes to **stdout** (markdown by default, JSON

Construct, review, score, and rewrite written content so it reads like a specific human wrote it, not an AI. Use this skill whenever the user wants to: write or draft prose for humans (docs, README, runbook, ADR, PR/commit message, blog post, LinkedIn post, email, Slack message, or a spoken explainer/tutorial video script); humanize or de-slop AI-generated text; check whether writing "sounds like AI"; review a draft for AI texture; rewrite content in their own voice; score a draft for authenticity or clarity; or tighten and sharpen prose. Also trigger on "humanize", "make it sound human", "sounds like AI", "does this sound like AI", "voice check", "review my draft", "rewrite in my voice", "tighten this up", "edit for clarity", "video script", "explainer script". Auto-detects content type and applies channel-specific rules. Defaults to a neutral, factual voice and never invents specifics to add texture. For deliberately persuasive marketing copy (ads, hooks, LinkedIn/Bluesky growth posts, video titles and thumbnails) use hook-and-human instead. For a CV/resume or a LinkedIn PROFILE - including de-slopping or humanizing one, or rewriting a headline, About section or experience bullets - use cv-and-human instead.

Tailor and polish a CV/resume - or a LinkedIn profile - to maximise its score in applicant tracking systems (ATS) and LinkedIn recruiter search, covering both keyword/JD-matching screeners and modern LLM-rubric scorers (which often take no job description at all). Use this whenever the user wants to optimise, tailor, rewrite, or "ATS-proof" a CV or resume, asks why their resume isn't getting past the ATS or automated screening, wants a keyword/gap analysis against a job description, wants to remove AI-generated texture from / de-slop / humanize a CV, resume or LinkedIn profile (a CV is its own channel with its own de-slop rules), or wants to check whether a CV parses cleanly. ALSO use for the LinkedIn PROFILE as a career document: "optimise my LinkedIn profile", "rewrite my LinkedIn headline", "my LinkedIn About section", "the experience bullets on my LinkedIn profile", "why am I not showing up in recruiter searches", "make my LinkedIn match my CV", "LinkedIn keywords", "personal brand positioning on my LinkedIn profile". Trigger even if there is no job description, as long as the user names something to be DONE to the document - tailored, ATS-proofed, parse-checked, de-slopped, or a specific LinkedIn field rewritten. Do NOT use for writing a LinkedIn POST or other social content - that is hook-and-human (persuasive) or clear-and-human (neutral). Do NOT use when the user has no target role in mind and is asking an open question about themselves - what roles they are credible for, what evidence they are missing, whether they are pigeonholed, or "does this look OK" over a bare CV - that is cv-evidence-base, which recovers the material this skill then tailors. Do NOT use for writing a CV from a blank page with no material to work from, or for generic career advice unrelated to a document.

Interrogates a CV or resume to recover the evidence that never made it onto the page, and works out which roles the person is genuinely credible for. Grades their background against role archetypes derived from what they actually did rather than from their job titles - including archetypes they are NOT credible for - asks the oblique questions that dislodge buried achievements, names where the page would get binned in a seven-second scan, and maintains an evidence base plus an action ledger across sessions. Use this whenever the user has NO target role in mind and is asking an open question about themselves: what roles they could realistically go for, whether they are pitching themselves right, whether they are pigeonholed, why they get no callbacks or interviews, what they are missing, or what their CV says about them. Use it when they hand over a CV with no clear instruction - "does this look OK", "be honest with me", "am I positioned right". ALSO use for a career change, a step up, going freelance or contracting, or restarting a job search after a long gap. Do NOT use when the user names something to be DONE to the document - tailor it to this job description, ATS-proof it, keyword/gap analysis, check it parses, de-slop it, or anything about a LinkedIn profile's headline, About section or recruiter-search keywords. That is cv-and-human, which is the right next step once this skill has established what is true and worth claiming.

Write, punch up, and review marketing copy that stops the scroll and converts, while still sounding like a real person. Use this skill whenever the user wants to: write a LinkedIn post, X/Twitter post, Bluesky post, cold or marketing email, ad, landing-page section, headline, hook, lead magnet, or YouTube packaging (video title, thumbnail text, description hook, end-screen/CTA, community post); "punch up" or "make this convert" or "make this land"; sharpen a CTA; or review copy for stopping power and conversion. Also trigger on "hook", "scroll-stopper", "make it punchy", "marketing copy", "sales copy", "cold email", "ad copy", "landing page copy", "lead magnet", "bluesky", "youtube title", "video title", "thumbnail", "make this convert". Deploys persuasion patterns (hooks, frameworks, one strong CTA) on purpose, reads WRITING_CONTEXT.md for voice and approved facts, and never fabricates specifics. For neutral, credibility-first writing (docs, README, plain explanation, or a teaching/explainer video script) use clear-and-human instead. For the LinkedIn PROFILE itself (headline, About section, experience bullets, recruiter-search keywords) use cv-and-human instead.

Aggregate sanctioned public job feeds into one deduplicated local database, match postings against the user's career lanes, and render a filterable self-contained HTML report. Use whenever the user wants to search or monitor job boards - "find me remote platform engineering roles", "what jobs came up this week", "check the job boards", "aggregate job listings", "what's new since I last looked" - or wants a report of current openings across multiple sites. Covers eight documented JSON APIs and RSS feeds (Arbeitnow, Jobicy, Remotive, Remote OK, Working Nomads, 4 Day Week, We Work Remotely, Python.org Jobs), weighted towards the German and EU-remote market. NOT for LinkedIn - that needs a signed-in session and is a separate tool. This skill never scrapes HTML, never works around a block, and never republishes what it collects.

The job is almost always the same shape: **data in → one HTML file out**, rendered by a template, viewable by double-clicking with no server. Keep that shape and the work stays small and reproducible. The two failure modes to design against are (1) non-determinism (charts that fetch live, timestam

Turn a loose trip idea into a plan someone can actually act on: a paced day-by-day itinerary and a budget that's been reconciled against what the traveler can spend. The output is a single editable Markdown file they can tweak by hand afterward.

`linearis` is a Node CLI for Linear.app — JSON output, smart ID resolution, cursor pagination, built for LLM agents. It ships two identical binaries, `linear` and `linearis`. It is **not** an MCP: no tool schemas land in context, so every fresh session pays a discovery tax. This skill pays that ta