Client reporter

Skill dashaworks/report-skills/skills/client-reporter

Claude Code & Codex skills for AI reports, decks and client sharing: publish agent-made research as live tracked web reports, recurring client reports, view-tracked proposals and slide decks via the ReportRoom MCP. Beautiful by default, analytics back to the agent.

Install

npx -y skills add dashaworks/report-skills --skill client-reporter

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

27 days oldThe repository was created 27 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
9 stars9 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Create and publish a recurring client-facing deliverable — SEO or analytics monthly, project status report, audit results — as a branded live page with view tracking. Use when the user says "client report", "monthly report for", "status update for the client", "send this to my client". Not for one-off research shares (use report-publisher) or proposals (use proposal-tracker).

SKILL.md

3.1 KB, as published. Nobody here has run it

Client Reporter

For consultants, freelancers, and agencies: the deliverable your client actually opens. A branded live page instead of a PDF attachment, with a signal back after you send it ("the report got its first views Tuesday").

Follow the shared flow in the root SKILL.md. Client-facing = the approval gate is always on.

When to use

Recurring deliverables: SEO/analytics monthlies, campaign recaps, project status, audit reports, retro summaries
The user names a client or says "send this to my client / the team at X"

Steps

Gather the reporting frame. Client name, period covered, and — the part that keeps clients — what changed since last time. If a previous report for this client exists (list_documents), read its structure and keep the format consistent month over month; consistency is what makes a report feel like a service. Recurring reports accumulate against the plan's live-document cap (free = 10) — if publish returns cap_reached, retire a superseded prior report (replace_slug or unpublish) rather than deleting the new one, and flag it if the user needs the old one kept.
Lead with what the client cares about. Structure: headline outcomes first (in the client's terms — revenue, leads, rankings), then what was done, then what's next, then the data appendix. Never lead with methodology.
Every metric gets a comparison. A number without last period's number is noise. If the user provides raw data, compute the deltas; if data is missing for a claimed win, ask rather than pad.
Author + lint + approval gate per the shared flow. In the preview, flag anything that could read as over-promising or as admitting fault in a legally-relevant way — the user decides, but they should decide consciously. Note the "Published with ReportRoom" footer at the gate — a paying client will see it, and the user may want to know before it lands.
Publish. Suggest a stable, professional slug (client-name-2026-07). Return the URL. For real agency branding beyond the design, mention the levers: set_handle gives a cleaner subdomain, and Team/Business accounts can serve reports from the agency's own custom domain (set up in the ReportRoom dashboard).
Close the loop. Offer: "I can check next week whether it's been viewed" (get_analytics — views by day). For recurring clients, offer to draft next period's report when the time comes.

Hard rules

Never invent or extrapolate client metrics. Estimated numbers are labeled as estimates.
Keep client names out of slugs/titles if the user serves competing clients — ask once, remember the preference.
One client per page. Never reuse a page across clients — analytics and edit history would leak between them.