
I help agency leadership teams align on growth priorities and build the execution capacity to make them move.
Gary Magnone has spent 15+ years on the agency side across strategy, delivery, accounts, operations and growth, and now works as a fractional operator embedded with agency leadership teams. His skills encode how he runs the conversations that turn a vague "delivery is painful" into one quantified problem, a price the owner has already justified, and a booked next step.
One command for Claude Code, a prompt for Claude or ChatGPT, or a ZIP of every skill folder.
Help me install the 5 skills by Gary Magnone from this GitHub URL in Claude: https://github.com/agiea-brain/agency-skills/tree/main/skills/gary-magnone Prepare only the skills in that folder, not other skills in the repository. Inspect each complete skill folder, not just SKILL.md. Preserve the original instructions and include all supporting files (such as references, scripts, templates and assets), keeping their relative paths intact. Package each skill as a downloadable ZIP containing its skill folder with SKILL.md directly inside that folder. If a skill only contains SKILL.md, package that folder too. Give me the download link for each ZIP and walk me through uploading and enabling it under Settings, Capabilities, Skills in Claude. If you cannot access GitHub or create downloadable files here, explain what I need to download and how to package and upload it myself. Do not invent missing files or claim the skill is installed before I upload it.
Paste this prompt into Claude. It packages each skill folder as a ZIP, supporting files included, and walks you through uploading them under Settings, Capabilities, Skills.
Turns client or stakeholder feedback on something already built (a staging page, a deck, a PDF, a mockup) into a numbered change note for the person who builds it (designer, developer, editor). Reads the feedback and the live thing, sorts each comment by what it changes for the user to confirm, finds every other place the same line or number appears, and writes "Hey [builder], [client] went through [the thing]..." with exact old and new copy, grouped by asset, and a closing ask for timing. Use when the user says "/direction", "translate this into direction for [builder]", "help me pass this along to [builder]", "what do I need to pass along to the designer", "write this as a note to [builder]", "[client] also sent this", "create a task for [builder] for [person]'s request", or pastes a client email of revisions and asks what changes. Never sends anything.
Use this skill when you have 30 minutes with the owner of a professional service firm or agency (an existing customer, a closed-lost contact, or a net new prospect) to find the one client-delivery problem worth building a custom solution or pilot for. Produces a quantified pain (annual dollar figure plus opportunity cost in the owner's own words), a price framing delivered on the call, and a booked next step. Also use for "delivery audit", "AI audit call", "discovery call for a custom build", "scope a pilot", "find the gnarliest workflow", "price on the call".
Writes the short note that goes out when someone hands a finished piece of work to a client, partner, or teammate (a deck, a tool, a strategy doc, a spreadsheet, a package of files). Opens the deliverable, pulls the last call with the person getting it, then drafts "Hey [name]! Here's [the thing]. Quick rundown:" with four to six plain bullets and a casual close, plus a short "before you send" list of anything sensitive. Use when the user says "/rundown", "bullets I can send along to [person]", "as I pass this over to them", "give them the rundown on what it is and how to use it", "a handful of bullet points to deliver with it", "tldr I can send via Slack", "brief note to go with this", or "match the bullets you wrote for the last one". Never sends anything.
Audits a finished deliverable claim by claim before it goes to a client, and traces a number back to its origin when somebody on the client side says it is wrong. Lists every factual claim in the deck, doc, or sheet, ties each one to a named primary source, and marks the ones that cannot be sourced so they get cut or backed up before the thing ships. Use when the user says "/source-check", "fact check all of this", "fact check yourself", "where did we get [the number]", "how do you know that", "I need to be 100% confident in the source for every finding", "read the sources before quoting any of it", "before this goes to [client]", or when someone on the client side says a figure in a deliverable is wrong. Never sends anything.
Reviews something a teammate or vendor built for a client (a page on staging or live, a dashboard, a PDF, a report, a page revamp) and writes the feedback note that goes back to them. Opens the real thing first, not a summary of it, checks it against the plan it was built from, splits the read into "the good" and "opportunities" ranked high, medium, and low, keeps it at the strategy level, and frames it so nobody on the team feels graded. Use when the user says "/work-review", "[person] is asking me to review [the thing]", "draft of the new [page]. thoughts?", "what's good, what's not so good, any high priority items to flag", "check out [person]'s updated dashboard", "is this page organized the way we planned", or pastes a teammate's handoff and asks what to say. Never sends anything.