Ejecuta cualquier Skill en Manus
con un clic

Ejecuta cualquier Skill en Manus con un clic

sales-phantombuster

Estrellas58

Forks7

Actualizado28 de junio de 2026, 12:09

PhantomBuster (phantombuster.com) platform help — code-free web/LinkedIn automation + scraping + lead-gen: 'Phantoms' (LinkedIn search export, scrapers, auto-connect, email finder), Flows that chain them, AI Agents + a cloud LinkedIn Leads database. Developer surface: a REST API (base phantombuster.com/api/v2, X-Phantombuster-Key-1 header) to launch agents (POST /agent/{id}/launch → container id) and fetch run results (/containers/fetch-result-object), plus the Leads storage and per-agent webhooks (secret param, 11s timeout) on run completion. Billed by execution time + slots; API on all paid plans. Use when launching a Phantom or fetching its scraped output via the API, wiring a run-finished webhook, handling the async agent→container model, managing slots/execution-time, or LinkedIn cookie/safety issues. Do NOT use for prospect-list / lead-sourcing strategy across tools (use /sales-prospect-list), enrichment strategy across vendors (use /sales-enrich), or outbound cadence strategy (use /sales-cadence).

Instalación

Instalar con Codex o Claude Copia este prompt, pégalo en Codex, Claude u otro asistente, y deja que revise la página de la skill y la instale por ti.

Ejecutar en Manus

Fuente

sales-skills

sales-skills/sales

Abrir repositorio de GitHub Ver repositorios del creador

Descarga

Ejecutar en Manus

Explorador de archivos

4 archivos

SKILL.md

readonly

Más de este repositorio

mismo repositorio

sales-do

sales-skills/sales

Routes any sales, marketing, ad, or GTM objective to the right specialized skill and outputs the install command for that skill plus a ready-to-paste prompt packed with the user's context. Asks clarifying questions when the objective is ambiguous, then hands back a copy-paste-runnable next step. Covers prospecting, outbound cadences, deals, proposals, forecasting, deliverability, enrichment, intent, content, coaching, CRO, SEO, launch directories, newsletters, email/SMS/push marketing, chatbots, influencer marketing, social media, employee advocacy, media relations, reviews, data hygiene, B2B advertising, retargeting, affiliate, loyalty, digital products, memberships, webinars, checkout, and platform-specific help. Use when the user has a sales or marketing question and isn't sure which skill to use, or wants a multi-skill sequence with a batch install command. Do NOT use to solve problems directly — this skill only routes.

2026-06-2858

sales-prospect-list

sales-skills/sales

Builds targeted B2B prospect lists for outbound campaigns from database providers (Apollo, ZoomInfo, Clay, Cognism). Use when you don't know who to target, your outbound list is too broad and response rates are low, you can't find the right decision-makers at target accounts, your ICP is vague, or your prospect list is full of bad-fit contacts. Do NOT use for enriching existing contacts with emails/phones (use /sales-enrich), interpreting buying signals (use /sales-intent), general Apollo platform help (use /sales-apollo), or hyperlocal / door-to-door / territory / route-based outbound to local SMBs that aren't on databases — restaurants, gas stations, salons, contractors, HVAC — (use /sales-field-sales).

2026-06-2858

sales-intent

sales-skills/sales

Interprets buying signals and prioritize accounts for outreach. Use when drowning in intent data and not sure what matters, can't tell which accounts to prioritize, buying signals aren't translating to pipeline, job change alerts are noisy, intent topics feel too broad, lead scores don't match actual deal quality, or not sure how to act on signals in outreach. Do NOT use for building prospect lists (use /sales-prospect-list), enriching contacts (use /sales-enrich), ZoomInfo-specific intent config (use /sales-zoominfo), or general Apollo platform help (use /sales-apollo).

2026-06-2858

sales-syncgtm

sales-skills/sales

SyncGTM (syncgtm.com) platform help — AI-native GTM data engine that unifies buying signals (job changes, funding, hiring, LinkedIn posts, web-traffic, custom AI signals), waterfall enrichment (verified emails/mobiles, 40+ providers), AI lead scoring (0–100 vs ICP), AI research agents, and a no-code Workflow Builder. Developer surface: an MCP server (api.syncgtm.com/mcp, OAuth browser sign-in, free-plan — works with Claude Code/Cursor to enrich, find emails/phones, read signals), a REST API (api.syncgtm.com, API key), webhooks, and 40+ integrations (HubSpot/Salesforce/Slack/Attio). Use when connecting the MCP server to an AI agent, enriching contacts via the API, acting on buyer signals, scoring leads vs an ICP, building a signal→enrich→outreach workflow, or choosing SyncGTM vs Clay/Trigify. Do NOT use for intent-data strategy across tools (use /sales-intent), prospect-list building (use /sales-prospect-list), or enrichment strategy across vendors (use /sales-enrich).

2026-06-2858

sales-cadence

sales-skills/sales

Designs multi-channel outbound cadences with timing, A/B testing, and content. Use when outbound sequences getting low reply rates, unsure how many touchpoints or what spacing to use, cadence emails sound generic, prospects ignoring calls, A/B tests not moving the needle, sequence performance plateauing, or not sure which channels to include in outreach. Do NOT use for marketing cold email (use /cold-email), automated email flows (use /email-sequence), or general Salesloft platform help (use /sales-salesloft).

2026-06-2858

sales-lagrowthmachine

sales-skills/sales

La Growth Machine (lagrowthmachine.com) platform help — multichannel B2B outbound automation across LinkedIn, email, and X: sequenced campaigns (conditions, branches, A-B), per-identity safety limits, waterfall lead enrichment, lookalike + LinkedIn intent signals, a shared inbox, AI copy + AI voice, and outcome tracking. Developer surface: REST API (apiv2.lagrowthmachine.com/flow, apikey query param) to enroll leads into audiences (POST /flow/leads), list campaigns/audiences/identities, and create inbox webhooks for real-time reply events; native HubSpot/Clay/Make/Slack, Zapier/n8n, and an MCP server — API/webhooks gated to the Ultimate plan. Use when enrolling leads via the API, wiring inbox webhooks (create a CRM contact on reply), managing identities/safety limits, building a multichannel sequence, or choosing LGM vs lemlist. Do NOT use for outbound cadence strategy across tools (use /sales-cadence), email deliverability/warmup (use /sales-deliverability), or generic iPaaS wiring (use /sales-integration).

2026-06-2858

name	sales-phantombuster
description	PhantomBuster (phantombuster.com) platform help — code-free web/LinkedIn automation + scraping + lead-gen: 'Phantoms' (LinkedIn search export, scrapers, auto-connect, email finder), Flows that chain them, AI Agents + a cloud LinkedIn Leads database. Developer surface: a REST API (base phantombuster.com/api/v2, X-Phantombuster-Key-1 header) to launch agents (POST /agent/{id}/launch → container id) and fetch run results (/containers/fetch-result-object), plus the Leads storage and per-agent webhooks (secret param, 11s timeout) on run completion. Billed by execution time + slots; API on all paid plans. Use when launching a Phantom or fetching its scraped output via the API, wiring a run-finished webhook, handling the async agent→container model, managing slots/execution-time, or LinkedIn cookie/safety issues. Do NOT use for prospect-list / lead-sourcing strategy across tools (use /sales-prospect-list), enrichment strategy across vendors (use /sales-enrich), or outbound cadence strategy (use /sales-cadence).
argument-hint	[describe what you need help with in PhantomBuster]
license	MIT
version	1.0.0
tags	["sales","prospecting","automation","platform"]
github	https://github.com/phantombuster

PhantomBuster Platform Help

Step 1 — Gather context

If references/learnings.md exists, read it first for accumulated platform knowledge.

What are you trying to do?
- A) Launch a Phantom via the API and fetch its scraped output (agent → container → result)
- B) Get notified when a run finishes (per-agent webhook)
- C) Pick/configure a Phantom (LinkedIn search export, profile scraper, auto-connect, email finder)
- D) Chain Phantoms into a Flow/Workflow, or add an AI Agent step
- E) Manage slots + execution-time limits, or the LinkedIn Leads storage
- F) Handle LinkedIn session cookies / avoid getting an account flagged
API or no-code? Code → REST API (X-Phantombuster-Key-1) + webhooks. No endpoint → run Phantoms in-app, or Zapier/Make/n8n.

Skip-ahead rule: if the user's prompt already provides enough context, skip to Step 2.

Step 2 — Route or answer directly

If the question is about...	Route to...
Prospect-list / lead-sourcing strategy across tools (which scraper, ICP)	`/sales-prospect-list {question}`
Contact enrichment strategy across vendors	`/sales-enrich {question}`
Outbound sequence/cadence for the leads you scrape	`/sales-cadence {question}`
LinkedIn outreach strategy / account safety across tools	`/sales-linkedin {question}`

When routing, give the exact command, e.g. "This is a strategy question — run: /sales-prospect-list build a LinkedIn ICP list safely".

Step 3 — PhantomBuster platform reference

Read references/platform-guide.md for the full reference — the Phantom/Flow/Agent model, the slots + execution-time + email-credits limits, the LinkedIn-cookie + safety model, plan tiers, and quick-start recipes (launch a Phantom; fetch results; wire a webhook).

Read references/phantombuster-api-reference.md for the integration surface — base https://phantombuster.com/api/v2, X-Phantombuster-Key-1 auth, the Agents (/agents/fetch-all, /agent/{id}/launch, /agent/{id}/fetch-output), Containers (/containers/fetch, /containers/fetch-result-object), and Storage (LinkedIn Leads DB) endpoints, JSend responses, the Agent-ID vs Container-ID model, and per-agent webhooks (secret query param, 11s timeout).

Answer using only the relevant section. Don't dump the full reference.

Step 4 — Actionable guidance

Focus on the user's specific situation:

Agent → container → result. POST /agent/{id}/launch queues a run and returns a container id; the scraped data isn't ready until that container finishes. Pick the output param deliberately, then read /containers/fetch-result-object?id=… for the structured leads.
Use a webhook, not tight polling. Set the per-agent webhook (Advanced Notification Settings). It POSTs on completion — but verify the secret query param (webhooks can't send custom headers) and return 200 within 11s (ack fast, process async) or it's treated as an error.
Auth is the X-Phantombuster-Key-1 header. From Workspace settings, shown once. Anyone with it can launch your agents — keep it server-side, rotate if leaked.
Mind slots + execution time. Phantom slots = concurrent automations (5/15/50 by tier); execution time = monthly run-hours (20/80/300). Long scrapes eat hours; you can't run more concurrent Phantoms than your slots. API is on all paid plans.
Scraping runs as you (session cookie). Phantoms use your LinkedIn li_at cookie — runs act as your account, so respect human-like limits to avoid restriction, and refresh the cookie when it expires. For account-safety strategy, route to /sales-linkedin.
It's a scraper/automator, not a CRM/sequencer. Pipe the scraped leads to your CRM/sequencer (or Clay) — for outreach design use /sales-cadence.

If you discover a gotcha, workaround, or tip not covered in references/learnings.md, append it there.

Gotchas

Best-effort from research (2026-06) — API verbatim from the docs; pricing/features from support + reviews. Confirm in-account.

Async: launch ≠ results. /agent/{id}/launch returns a container id immediately; the data lands when the container finishes — webhook or poll /containers/fetch, then read fetch-result-object.
Webhooks use a secret query param + have an 11s timeout. No custom headers → verify the secret, not a header signature; respond 200 in <11s or it's a server error (ack then process async).
X-Phantombuster-Key-1 is the auth header. Not Authorization: Bearer. Key from Workspace settings, shown once; rotate if leaked.
v2, milliseconds. v1 timestamps are seconds, v2 are ms — use v2 and don't mix.
Session-cookie scraping has account risk. Phantoms act as your logged-in account (LinkedIn li_at); over-aggressive runs can get the account restricted, and cookies expire. Keep volumes human, refresh cookies.
Slots + execution-time cap throughput. You can't exceed concurrent slots, and long scrapes burn monthly execution hours — size the plan to volume.
Compliance gray area. Scraping LinkedIn/sites can violate their ToS; use responsibly and within data-protection law.

Related skills

/sales-prospect-list — Prospect-list / lead-sourcing strategy across tools (PhantomBuster is one of the scrapers) — which tool, ICP, list hygiene
/sales-enrich — Contact/account enrichment strategy across vendors (pair scraped leads with enrichment)
/sales-cadence — Outbound sequence strategy for the leads you scrape
/sales-linkedin — LinkedIn outreach / account-safety strategy across tools
/sales-do — Not sure which skill to use? The router matches any sales objective to the right skill. Install: npx skills add sales-skills/sales --skill sales-do -a claude-code

Examples

Example 1: Launch a Phantom and fetch its scraped leads via the API (developer/automation)

User says: "How do I run my LinkedIn Search Export Phantom from my backend and get the results?" Skill does: Walks Recipe 1 — POST https://phantombuster.com/api/v2/agent/{id}/launch with the X-Phantombuster-Key-1 header (and a chosen output), which returns a container id; then poll /containers/fetch?id=… until done and read /containers/fetch-result-object?id=… for the structured leads (JSend data). Recommends a webhook instead of tight polling, and keeping the key server-side. Result: Backend launches the scrape and ingests the leads reliably.

Example 2: Get notified when a run finishes (webhook)

User says: "I want my system pinged when a PhantomBuster run completes." Skill does: Sets the per-agent webhook in Advanced Notification Settings, configures a secret (since webhooks can't send custom headers → verify the secret query param), and stresses the 11-second timeout — the handler must return 200 fast and process async, then call fetch-result-object for the data. Offers Zapier/Make as a no-code alternative. Result: Run-completion events drive ingestion without polling.

Example 3: My LinkedIn account got a warning from PhantomBuster

User says: "LinkedIn flagged my account after running Phantoms — what happened?" Skill does: Explains Phantoms run as your account via your session cookie, so aggressive volumes (connects/messages/scrapes) trigger LinkedIn's limits. Recommends staying within human-like daily caps, spacing launches, not maxing connection requests, refreshing the cookie when it expires, and warming new accounts. Routes account-safety strategy: "run: /sales-linkedin safe LinkedIn automation limits." Result: Safer automation settings and a recovery plan.

Troubleshooting

My launch succeeds but I get no data

Symptom: /agent/{id}/launch returns success but the results are empty/missing. Cause: The run is asynchronous — launch only queues it and returns a container id; the data isn't ready until the container finishes (and you must read the result object, not the launch response). Solution: Take the containerId from the launch response, poll /containers/fetch?id=… until it's finished (or use a webhook), then fetch /containers/fetch-result-object?id=… for the structured leads. Make sure the Phantom's output setting actually produces a result object.

My webhook isn't firing / times out

Symptom: The completion webhook never arrives or errors. Cause: The URL wasn't set in Advanced Notification Settings, the handler took >11 seconds (cancelled → server error), or you're checking for a header signature that doesn't exist. Solution: Set the agent's webhook URL, return 200 within 11s (ack immediately, do work async), and verify the secret query parameter (webhooks can't send custom request headers). Re-test by launching the agent.

I've run out of execution time / slots

Symptom: Phantoms won't launch or queue up. Cause: You hit your monthly execution-time hours or your Phantom slot (concurrency) cap. Solution: Check usage in the dashboard; reduce scrape scope (fewer profiles, narrower search), stagger launches, and free up slots by pausing idle Phantoms. Upgrade the tier if volume genuinely needs more hours/slots. For list-building strategy that minimizes wasted runs, use /sales-prospect-list.