Deepline Quickstart

Deepline Quickstart

Run a high-confidence demo recipe to show the user what Deepline can do. Pick the most relevant recipe below, or default to Recipe 1 if no context is given.

Always prefer the hardcoded recipes below. /deepline-gtm is always available as a fallback but should only be used if: (a) a recipe command fails and all fallbacks are exhausted, or (b) the user's ask doesn't match any recipe here. Never invoke it preemptively.

Execution flow

Follow this pattern for every recipe:

1. Tell the user what you're about to do — explain the goal and which data source(s) you'll use, before running anything.
2. Register a session start with deepline session start --steps '[...]' matching the recipe steps. If you have the user's original request text, include it with --user-prompt "..." so opted-in prompt telemetry is preserved.
3. Run the recipe directly. For this default quickstart, do not spend time on per-step session status / session start --update chatter; one session start plus one output registration is enough.
4. Register output with deepline session output --csv --label "..." after any CSV is produced.
5. Tell the user the results — summarize what came back, where it came from, and what they can do next.

CLI surface

This quickstart needs to be fast. Do not run deepline --version, deepline auth status, or separate CLI discovery commands. The fast path performs one inline deepline enrich --help check to choose the compatible command syntax. If the installed CLI is SDK/V2, use --name quickstart-ny-cto-email and the hyphenated person-linkedin-to-email prebuilt id. If the installed CLI is legacy/V1, omit --name and use person_linkedin_to_email_waterfall.

Session commands reference

deepline session start --steps '["Step 1", "Step 2"]' --user-prompt "Original user request"
deepline session start --update  --status running|completed|error|skipped
deepline session status --message "What's happening right now..."
deepline session output --csv  --label "Label for the table"
deepline session usage [--session-id UUID] [--json]

---

Recipe 1 — Find CTOs at NY startups

Goal: Find 5 CTOs at startups in New York with verified emails and LinkedIn profiles. Data sources: Dropleads (people search) + waterfall email enrichment via person-linkedin-to-email on SDK/V2 or person_linkedin_to_email_waterfall on legacy/V1.

Steps:

1. Search Dropleads for CTOs in New York
2. Waterfall enrich emails
3. Display results

Fast path

For the default quickstart, run this whole block as one Bash call. Do not split it into separate tool calls. Do not inspect the JSON, run csv show, print the CSV with Python, or run extra validation after deepline session output; those checks make the quickstart miss the one-minute budget.

set -e
mkdir -p deepline/data

deepline session start --steps '["Search Dropleads for CTOs in New York", "Waterfall enrich emails", "Display results"]' --user-prompt "Run the default Deepline quickstart demo"

deepline tools execute dropleads_search_people --json --payload '{
  "filters": {
    "jobTitles": ["CTO"],
    "personalStates": {"include": ["New York"]},
    "employeeRanges": ["1-10", "11-50", "51-200"]
  },
  "pagination": {"page": 1, "limit": 5}
}' > deepline/data/quickstart_search.json

python3 - &1 | grep -Eq -- '(^|[[:space:]])--name([[:space:] --output  --name quickstart-ny-cto-email --all \
  --with '{"alias":"email","tool":"person-linkedin-to-email","payload":{"linkedin_url":"{{linkedin_url}}"}}'

Legacy/V1:

deepline enrich --input  --output  --all \
  --with '{"alias":"email","tool":"person_linkedin_to_email_waterfall","payload":{"linkedin_url":"{{linkedin_url}}"}}'

Register the output CSV after this step.

Step 3 — Display results

After the fast path finishes, do not run another command just to display rows. Tell the user the enriched CSV path and that emails were filled via the dedicated LinkedIn-to-email waterfall. Mention they can go deeper — phone, firmographics, job change signals — with /deepline-gtm.

Fallback (if Step 1 errors)

Tell the user, then try Dropleads:

deepline tools execute dropleads_search_people --payload '{
  "filters": {
    "jobTitles": ["CTO", "Chief Technology Officer"],
    "personalCountries": {"include": ["United States"]},
    "personalStates": {"include": ["New York"]},
    "personalCities": {"include": ["New York"]}
  },
  "pagination": {
    "page": 1,
    "limit": 5
  }
}'

Last resort

If all commands fail, tell the user, then invoke /deepline-gtm:

> Find 5 CTOs at startups in New York with their emails and LinkedIn profiles.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: getaero-io
• Source: getaero-io/gtm-eng-skills
• License: MIT
• Homepage: https://code.deepline.com

Install and usage instructions live in the source repository linked above.