agentsclimarketplace
Sign in

Call prep

Skill naveedharri/benai-skills/shared-skills/call-prep

Prepare for sales calls and output a live, shareable BenAI-branded dashboard (not a chat brief). Researches every attendee and company via web, Attio CRM, Gmail, and Fireflies, then deploys a neo-brutalist "command center" dashboard to a live URL. Works for one call or a whole day. Trigger with "prep me for my call with [company]", "call prep", "prep all my calls today", "get me ready for [meeting]", or "/call-prep".From its SKILL.md

Install
npx -y skills add naveedharri/benai-skills --skill call-prep

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

SKILL.md

5.8 KB, ~1.4k tokens by cl100k_base, as published. Nobody here has run it

Call Prep → Live Dashboard

Prep any sales call (or a whole day of them) and hand back one shareable live dashboard URL. This is a hybrid: the research depth of call prep, rendered in the BenAI instant-ui design language. The dashboard is the deliverable, never a markdown brief pasted into chat.

[!important] Non-negotiable output Every run ends with a deployed, shareable dashboard URL in the BenAI design system defined in references/dashboard-template.html. Build by cloning that template, not from scratch. Report the URL first.

Workflow

1. Find the calls

  • If the user names a call/company, use that. If they say "today" / "my calls", pull the calendar (calendar connector or gws calendar) for that day and list every external meeting. Exclude internal syncs (team standups, 1:1s with colleagues).
  • Confirm scope only if genuinely ambiguous (e.g. many calls and unclear which). Otherwise proceed.

2. Research every call (parallel)

For multiple calls, fan out one sub-agent per call in parallel (Agent tool, general-purpose, run in background). Each agent researches its call and returns a compact brief. For a single call, do it inline.

Per call, pull from whatever is connected, and skip gracefully what is not:

  • Web + LinkedIn, who the person is, the company (what they do, size, industry, recent news/funding), attendee roles. (WebSearch / WebFetch)
  • Attio CRM, existing record, deal stage, budget band, notes. Load: ToolSearch "select:mcp__7320cfa0-48c2-49b9-8bd6-96fe9605dabd__search-records,...__list-records,...__list-lists".
  • Gmail, recent threads with the attendees/domain: open questions, commitments, whether a proposal or recap was sent. Use gws gmail search "<domain>" (Bash).
  • Fireflies, prior call transcript/recap if the meeting already ran. Load: ToolSearch "select:mcp__plugin_benai-suite_fireflies__fireflies_search,...__fireflies_get_transcript".

Each brief must classify the call's stage (see color map below) and surface any action flag (unsent proposal, save situation, no-show to acknowledge, qualification risk).

3. Build the dashboard

  • Read references/dashboard-template.html and clone it. It carries the full design system verbatim: cream canvas, 3px hard borders, 6px hard shadows, gradient top stripe, system font + SF Mono, weight-900 headings, and the correct BenAI logo.
  • Fill it in:
    • Header, title "Sales Call Prep", date chip. Correct logo only.
    • Hero, the boxed title + a one-line plain subtitle (date + "all times [TZ]") + a 4-up stat strip (calls today, in closing motion, saves needed, action items). No green-dot eyebrow. Ever.
    • Priorities strip, the 2-4 things to do before the calls (from the action flags). Skip the section only if there are genuinely none.
    • One card per call, clone the .call card. Each: time, stage badge, contact + company, a one-line status flag where relevant, the play, top 3 questions, 2-3 objections with responses. Same layout for one call (single card) or many (grid).
  • Stage color map (card top bar + badge): save = red, close / warm = green, discovery = blue, qualify / recovery = amber.

4. Deploy to a live URL

  • Save the file to ~/Desktop/builds/benai/call-prep-<date>.html.
  • Publish as an Artifact (default-private, safe for prospect PII, gives a claude.ai share link the user controls). Use Vercel only if the user explicitly asks for a public URL.
  • To update an existing day's dashboard, republish the same file path (keeps the URL).

5. Save the source + report

  • Save the compiled markdown behind the dashboard to Team/BenAI/Profiles/Aryan/Sales-OS/Calls/<date>-call-prep-pack.md (Obsidian syntax, wikilinks, frontmatter) as the durable record.
  • Lead the reply with the live URL, then a short priorities summary and an at-a-glance table.

Hard rules (from user feedback)

  1. The output is always a deployed dashboard URL, one call or many. Never stop at a chat brief.
  2. Use the real BenAI logo: a circle + a single U-shaped smile. No eyes, no wide grin. Markup is in the template; never alter it.
  3. No AI-tell hero eyebrow, never the ● NAME · DATE · META mono line with a colored dot. Put context in the header or subtitle instead.
  4. No em dashes in copy. No marketing cliches ("leverage", "unlock", "transform", "seamless", "cutting-edge"). Plain, specific, builder-voice.
  5. Headings weight 900. All numbers/times/money in SF Mono. Hard borders and shadows only, no rounded corners, no blur.
  6. Everything grounded, web, Attio, Gmail, Fireflies. If a source is empty, say "nothing found", never invent history.

This skill shares its design language with the instant-ui skill. If the brand system there changes, mirror it into references/dashboard-template.html.

Self-improvement

This skill is never finished. Improve it as you use it.

  • When the user corrects how a step was done, update the relevant reference file (or this SKILL.md) so the correction sticks. Do not just fix it for this run.
  • When a correction is a hard rule ("always do X", "never do Y"), add it as a permanent rule here.
  • When the user says an output was genuinely good, save it to references/examples/ so it becomes a model for future runs.
  • Keep the skill small while you do this: when you add something, run the deletion test and cut anything that no longer changes behavior.

What ships with it: 1 file

16.1 KB alongside SKILL.md

references/

Gives 0 of the 12 instructions most sales audience skills give in ~1.4k tokens

Counted across 401 of the 401 authors here whose files we hold, read 2026-08-07

  • Read product marketing context before asking questionsin 21 of 401, across 11 files
  • Acknowledge competitor strengths honestlyin 18 of 401, across 7 files
  • Start every page with a summaryin 15 of 401, across 4 files
  • Use a single, low-friction call to actionin 15 of 401, across 7 files
  • Create a single source of truth for each competitorin 14 of 401, across 3 files
  • Make each follow-up email add new valuein 11 of 401, across 5 files
  • Cut any sentence that does not drive a replyin 10 of 401, across 4 files
  • Tie personalization directly to the problemin 10 of 401, across 4 files
  • Write paragraph comparisons for each dimensionin 9 of 401, across 3 files
  • Link between related competitor pagesin 9 of 401, across 3 files
  • Keep subject lines short and lowercasein 9 of 401, across 3 files
  • Define ideal customer profile from top customersin 9 of 401, across 3 files

Said here and by no other author read

  • deploy a shareable live dashboard URL
  • report the live URL first
  • clone the dashboard template instead of building from scratch
  • fan out one sub-agent per call in parallel
  • exclude internal team syncs
  • save the compiled markdown as the durable record

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.