> ## Documentation Index > Fetch the complete documentation index at: https://docs.gravitygtm.com/llms.txt > Use this file to discover all available pages before exploring further. # Use Helium in Slack: profile, search, chaining, and network import > Use Helium in Slack via @mentions, DMs, or `/usehelium`: fundraising profile intake, investor/LP search chaining, network import, spend confirmations, and daily limits. This page covers what Helium can do once it's [installed and signed up](/helium/slack-install) in your workspace, and the guardrails that keep it predictable to use. ## Talking to it * **@mention** it in a channel: `@Helium ...` * **DM** it directly — no @mention needed * **Slash command**: `/usehelium ` * **Follow-ups in a thread** it's already replied in don't need a fresh @mention. ## Fundraising fit profile Before your first search, Helium has a short, natural conversation to collect your fit profile — it only asks for what's actually missing, never a fixed script: * **Raising equity from VCs/angels?** Stage, sector, ICP, geography, round size, known competitors. * **A fund raising from LPs?** Fund name, thesis, check size, target LP types (family offices, HNW individuals, recently-exited founders, institutional allocators, angels). Once set, Helium reuses it on every future search — it won't re-ask unless you say something implying it changed (e.g. "actually we're raising a Series A now"). ## Investor/LP research chaining Ask a compound question and Helium chains the underlying tool calls itself: > "Find seed fintech investors, then pull the full profile on the top 3." This runs a search → enrich chain automatically. Naming specific competitors (or having them in your profile) triggers a **competitor-driven discovery** fold-in — funds that backed those companies get surfaced and flagged "lookalike via {company}": > "Who funded Ramp and Ramp's competitors?" ## Network import Upload (or paste the contents of) your LinkedIn connections export — from LinkedIn's **Settings → Data privacy → Get a copy of your data → Connections** — and Helium imports it to power **warm-path matching**: future search results flag when someone in your own network already has a tie to a matched investor or fund. This is free and one-time; Helium will suggest it once but won't nag about it repeatedly. ## Lists and batches Paste a list of named investors/funds/companies and Helium researches each independently, then compiles **one combined answer** — never a reply per item. For a batch of more than roughly 5-10 items, it posts periodic progress updates rather than going quiet until the end. ## Honest answers, live cost transparency Helium never fabricates a fund name, investor name, check size, or portfolio company — an honest "not found" is preferred over a plausible-looking guess. Every answer that spent credits ends with exactly what it cost: ``` _Spent 12 credits · 988 remaining_ ``` ## Automated outreach sequences Connect your Gmail account first — go to your [dashboard](https://lite.gravitygtm.com/dashboard) and click **Connect Gmail** under the Helium section (one-time; Helium sends from your own address, never a shared one). Gravity's own shared Google app is still going through Google's verification review, so if your company uses Google Workspace, [setting up your own Google OAuth app](/helium/gmail-oauth-client-setup) first is recommended — it connects instantly, with no wait on that verification. This is a way to skip the wait today, not a permanent extra step; once Gravity's shared app is verified, plain **Connect Gmail** works directly for everyone. Not on Workspace? You may see Google's "app isn't verified" warning when connecting through the shared app in the meantime, or be unable to connect at all until Gravity adds your address to an interim allowlist — reach out if you hit this. Once connected, ask Helium to start a sequence to an investor or LP by email — for example: > start a sequence to [jane@examplevc.com](mailto:jane@examplevc.com) Helium schedules a 4-touch outreach sequence (an initial email, then follow-ups roughly 5, 9, and 14 days later) from your own connected Gmail account. It does **not** draft or send anything the moment you ask — a background process drafts each touch shortly before it's due, runs it through the same anti-hallucination checks as everything else Helium writes, and posts it to this Slack channel for your review: > **Touch 1 ready for Jane VC \<[jane@examplevc.com](mailto:jane@examplevc.com)>** > *Subject: ...* > *(drafted email body)* > \[✅ Approve & send] \[✏️ Edit] \[⏭️ Skip this touch] Nothing is ever sent without a human clicking **Approve & send** — this is a hard rule, not a suggestion the model can talk itself out of. **Edit** opens the draft in an editable form; your edited version still passes through the same safety checks before it can send. **Skip** moves on to the next touch in the cadence without sending this one. If the investor replies at any point, Helium detects it and automatically stops the sequence — no further touches are drafted or sent into a thread that's already gotten a response. The Slack message also classifies the reply (interested, wants to meet, pass, referral, or unclear) and adds a generic suggested next step, instead of just a plain "they replied" line — see [Reply classification](/helium/monitoring-and-discovery#reply-classification) for exactly how that works. Add your name and company (and a real scheduling link, if you have one) to your fundraising fit profile before starting sequences — Helium's safety checks use these: a draft that doesn't identify itself as you gets flagged for review instead of sent, and a scheduling offer only survives if it points to a scheduling link you've actually configured, never an invented one. ## Monitoring & discovery Once Gmail is connected, Helium also runs a few background checks with no extra setup: * **Calendar meeting-prep briefs** — once daily, Helium checks today's calendar for meetings with investors you've already started a sequence with, and posts what it knows about their fund before the call. * **Lookalike investor discovery** and **LinkedIn engagement monitoring** are opt-in — tell Helium something like "turn on lookalike monitoring, our competitors are Ramp and Mercury" or "start monitoring my LinkedIn engagement" and it should set the relevant profile fields for you via `set_fundraising_profile`. If it doesn't pick that up, ask it to call `set_fundraising_profile` explicitly, or set the fields directly with `PUT /v1/fundraising/profile` — see [Monitoring & discovery](/helium/monitoring-and-discovery) for the exact field names. Both run free on a schedule once opted in; `find_lookalike_investors` is also available as a paid on-demand tool if you want a lookalike search right now instead of waiting for the weekly check. See [Monitoring & discovery](/helium/monitoring-and-discovery) for exactly what each check looks at, how often it runs, and the full free-vs-billed breakdown. ## Spend confirmation Before running anything estimated above **50 credits or 10 items**, Helium pauses and posts a card instead of just running it: > **Confirm before I proceed** > *(plan summary + credit estimate)* > \[Run it] \[Cancel] This is a hard rule, not a suggestion the model can talk itself out of. ## Free-tier daily limit A free-trial workspace gets **20 agent messages per day** (resets at midnight UTC) — tracked **separately from Horizon's own daily limit**, so a workspace with both apps installed gets an independent budget for each. **A workspace on any paid credit pack is never capped** — only the request rate and credit balance shared with the REST API and MCP server (and with Horizon, if also installed) apply. See [Credits & Pricing](/helium/credits-and-pricing). ## Out of credits If a request runs out of credits mid-conversation, Helium says so plainly and includes a top-up link.