Community라이팅 & 에디팅github.com

outreachmagic/email-finder

Waterfall email enrichment through trykitt, Icypeas and MillionVerifier. Works standalone or with Outreach Magic for dedup and persistent storage.

지원 대상~Claude Code~Codex CLI~Cursor
npx skills add outreachmagic/email-finder

Ask in your favorite AI

Open a new chat with this agent skill pre-loaded.

문서

Email Finder

Find work emails when you have name + company domain. trykitt first, Icypeas on miss.

Works standalone. Just needs API keys. Pairs with Outreach Magic for credit-saving dedup, persistent storage, and cross-session availability.

Setup

Standalone (no OM)

Just API keys. Results print to stdout. No database needed.

KeyFor
TRYKITT_API_KEYtrykitt.ai (first in waterfall) — trykitt.ai
ICYPEAS_API_KEYIcypeas (fallback) — app.icypeas.com
python3 scripts/email_finder.py config  # verify keys loaded
python3 scripts/email_finder.py find --name "Jane Doe" --domain acme.com
# → prints result to stdout

With Outreach Magic (dedup + save)

Adds pre-flight dedup (skip leads already in OM) and saves results to your local SQLite pipeline. Requires outreachmagic skill with pipeline.py login.

KeysFor
All standalone keys above +
OUTREACHMAGIC_AGENT_KEYOM login via pipeline.py login
python3 scripts/email_finder.py find --name "Jane Doe" --domain acme.com --save --workspace CLIENT
python3 scripts/email_finder.py batch-find --workspace CLIENT --yes --workers 3 --delay 3 input.json

Before find/batch with OM, confirm keys: python3 <SKILLS>/outreachmagic/scripts/pipeline.py sync-secrets --check --json or python3 scripts/email_finder.py config.

Keys sync via Dashboard → pipeline.py sync-secrets. Verify source: python3 scripts/email_finder.py config (*_api_key_source should be agent_secrets).

Batch format

[{"lead_id": 12345, "name": "Jane Doe", "company_domain": "acme.com"}]

Production batch defaults

ModeFlags
Waterfall--workers 3 --delay 3
IcyPeas only--workers 2 --delay 3
TryKitt only--workers 3 (optional --delay 0.2)

Agent rules

  1. With OM: check before find (check / find with --save). Standalone: skip check — run find directly.
  2. Never fabricate emails.
  3. Waterfall: trykitt → Icypeas when both keys set.
  4. With OM: tags trykitt_attempted / icypeas_attempted; mv_attempted after MillionVerifier bulk (result lives in OM email_verification_status). Found state is leads.email, latest_source, and email_verification_status.
  5. With OM: lead_id on every row; --workspace required for OM save. Standalone: omit --save or use --no-save to print results to stdout.
  6. Run batch-find --dry-run before --yes to see skip counts (leads may already have email in OM while CSV email is empty).
  7. batch-find re-checks OM immediately before each API call (skips leads resolved since batch start).
  8. With OM: batch-find writes CSV/JSON under outreachmagic/exports/, then saves to OM. Standalone: add --skip-om to run without OM (writes to cwd).
  9. COMPLETE box shows IMPORT and RELAY (pending snapshots — run pipeline.py sync; upload is never automatic).
  10. Credits1 credit per email found (trykitt / Icypeas) or 1 credit per email verified (MillionVerifier). Not-found lookups cost 0 credits.

Batch input

[{"lead_id": 12345, "name": "Jane Doe", "company_domain": "acme.com"}]

Optionally include "linkedin_url" when available — trykitt uses it for higher match rates.

Commands

# Standalone (no OM) — prints result to stdout
python3 scripts/email_finder.py find --name "Jane Doe" --domain acme.com

# Standalone batch (no OM) — writes CSV to cwd
python3 scripts/email_finder.py batch-find --skip-om --yes --dry-run input.json

# With OM — find + dedup + save
python3 scripts/email_finder.py find --name "Jane Doe" --domain acme.com --save --workspace CLIENT

# Dry run (skip counts, no API spend)
python3 scripts/email_finder.py batch-find --workspace CLIENT --dry-run outreachmagic/batches/leads.json

# Batch (find + OM save)
python3 scripts/email_finder.py batch-find --workspace CLIENT --yes \
  --output-base outreachmagic/exports/emails --workers 3 --delay 3 outreachmagic/batches/leads.json

# OM save only — after failed import or --no-save run (accepts batch .csv or .json)
python3 scripts/email_finder.py import-to-om --file outreachmagic/exports/emails.csv --workspace CLIENT

python3 scripts/email_finder.py update --check

# MillionVerifier (optional)
python3 scripts/email_finder.py config
python3 scripts/email_finder.py verify-credits
python3 scripts/email_finder.py verify-bulk --workspace CLIENT --dry-run
python3 scripts/email_finder.py verify-bulk --workspace CLIENT --poll --yes

MILLIONVERIFIER_API_KEY in a local .env may show ***; OM agent_secrets.env overrides via ensure_env_loaded().

Resume a crashed batch by re-running the same batch-find command (skips completed API rows). If a run failed with network/auth errors, use --retry-errors to re-attempt errored rows without deleting the checkpoint.

Common workflows

User saysYou do
"Find Patrick at stripe.com"find --name … --domain stripe.com (standalone) or with --save --workspace W (OM)
"Find emails for my CSV" (with OM)batch-find --dry-runbatch-find --yes
"Find emails for my CSV" (standalone)batch-find --skip-om --dry-runbatch-find --skip-om --yes
"Retry failed email lookup"Same batch-find command with --retry-errors

Troubleshooting

  • ModuleNotFoundError: data_freshness — run pipeline.py update on outreachmagic.
  • COMPLETE shows ⚠ No import in IMPORT section — results are on disk; import-to-om --file {output-base}.csv --workspace W
  • CSV has emails, OM empty — batch save failed; import-to-om --file {output-base}.csv --workspace W
  • import-profiles timed out — results are on disk; use import-to-om or re-run with smaller batches.
  • IcyPeas ~10% hit rate — poll timeout; raise icypeas_poll_attempts in config
  • Checkpoint skipped everything after errors — re-run with --retry-errors, or delete {output-base}.csv / .json and start fresh.
  • --provider icypeas with no key — fails fast; add key at app.outreachmagic.io → Settings, then sync-secrets --check

Funnel

Starts useful alone: find / batch-find --skip-om. Pairs with lead-enrich (research → domain) and Outreach Magic (credit-saving dedup, persistent SQLite, cross-session availability). Learn more at outreachmagic.io. Both companions skip leads already tagged (serper_attempted / trykitt_attempted / icypeas_attempted).

관련 스킬

thananon/management-talk

Rewrite engineer-to-engineer content for engineering-org leadership (VPs, directors, PMs, release managers, execs in an engineering-savvy company) and shape it for the channel it is going to — JIRA comment, Slack post, async standup line, email, or meeting talking-points. Trigger when the user asks to write/rewrite for management / exec / VP / director / PM / release manager, asks for an "executive summary / leadership update / status update", says "make this less technical / less jargony", or asks for a slack / email / standup / meeting version of work originally written engineer-to-engineer.

community

vinodonweb/job-application-answers

A claude skills that helps you to write job application questions.

community

eronred/app-marketing-context

When the user wants to create or update their app marketing context document. Also use when the user mentions "app context", "marketing brief", "app positioning", or when starting any ASO or app marketing project. This is the foundation skill — all other skills check for this context first.

community

wordpress/wp-wpcli-and-ops

Use when working with WP-CLI (wp) for WordPress operations: safe search-replace, db export/import, plugin/theme/user/content management, cron, cache flushing, multisite, and scripting/automation with wp-cli.yml.

community

quickcall-dev/skills

Agent Skills for documentation, fleet orchestration, and parallel task execution — works with Claude Code, Cursor, Copilot, Gemini CLI, Codex, and 25+ agents

community

district-solutions/open-agent-tools-coder

Enables small-to-large self-hosted ai models to use local source code when running tool-calling agentic workloads. We actively data mine 20,900+ (2+ TB) popular github repos using large and small ai models to create reuseable: json, markdown and parquet files for local-first tool-calling models.

community