GTM Meta Skill

Use this skill for prospecting, account research, contact enrichment, verification, lead scoring, personalization, and campaign activation.

1) What this skill governs

• Route GTM decisions, safety gates, and provider/quality defaults before execution.
• Keep long command chains and tooling nuance in sub-docs; provider-specific implementation detail in provider-playbooks/*.md.
• Provide clear entry points for both paid and non-paid workflows, including --rows 0:1 one-row pilots.

Process/goal

Customer is generally trying to go from "I have an ICP" to "Here's a list of prospects with email/linkedin and very personalized content or signals". They may be anywhere in this process, but guide them along.

Discovery order: companies first, then people. When the task requires finding contacts at companies matching criteria (portfolio, ICP, hiring signal), discover the company set first, then find people at each company. Do not start with broad people-search queries.

Documentation hierarchy

• Level 1 (SKILL.md): decision model, guardrails, approval gates, links to sub-docs.
• Level 2 (phase docs): finding-companies-and-contacts.md, enriching-and-researching.md, writing-outreach.md, prompts.json.
• Level 2.5 (recipes/*.md): step-by-step playbooks for specific tasks (email lookup, LinkedIn resolution, waterfall patterns, contact finding, actor contracts). Search like code with Grep.
• Level 3 (provider-playbooks/*.md): provider-specific quirks, cost/quality notes, and fallback behavior.

No-loss rule: moved guidance remains fully documented at its canonical level and is linked from here.

CLI family fallback

If Deepline CLI V2 or SDK mode seems broken while running a GTM task, check deepline switch status. Use deepline switch sdk to move an installer-managed CLI to SDK mode, or deepline switch python to roll back to the Python CLI. Auth is host-scoped and should carry across both families.

2) Read behavior — MANDATORY before any execution

STOP. Do not call any provider, run any `deepline tools execute`, or write any search command until you have opened the correct sub-doc for your task.

These skill docs and sub-docs are not generic documentation — they are distilled from hundreds of real runs and encode exactly what works, what fails, and why. They contain validated parameter schemas, correct filter syntax, parallel execution patterns, tested sample payloads, and known pitfalls that took many iterations to discover. Think of them as shortcuts: reading a doc for 5 seconds saves you from 10 failed tool calls, wasted credits, and garbage output. Every time an agent skips reading the docs and tries to "figure it out" from first principles, it re-discovers the same failure modes that are already documented and solved.

SKILL.md is the routing layer — it tells you WHERE to go, not HOW to execute. The sub-docs and task-specific skills contain the HOW. Without them you will guess parameters, pick wrong providers, run searches sequentially instead of in parallel, and produce garbage results. This has happened repeatedly.

Open the right doc BEFORE executing

This is not optional. Read the matching doc. Do not skip this step. Do not "just try one provider real quick" or "just run one search to see." These docs exist because the correct approach was non-obvious and had to be learned through trial and error — they are shortcuts that let you skip straight to what works.

!important READING MULTIPLE DOCS IS A GREAT IDEA AND OFTEN SUPER ESSENTIAL. JUST READ MORE.

Routing rules — match your task to a doc and READ IT:

If you are hand-authoring enrich columns instead of using a native play, jump straight to the "Handmade step shape quick reference" section in enriching-and-researching.md. That section spells out the exact runtime contract for run_javascript, extract_js, result, and persisted matched_result.

Recipes: step-by-step playbooks for specific tasks (check before executing)

The recipes/ directory contains battle-tested playbooks. Before you start executing, scan this list and read any recipe that matches your task.

When a recipe matches: follow it step-by-step as your execution plan. Recipes encode hard-won sequencing and provider choices — trust them over generic guidance or your own intuition. If the user's request doesn't perfectly fit, adapt the recipe using the phase docs above, but keep the recipe's structure and ordering as your baseline.

If none match, grep for more specific keywords: Grep pattern="<keyword>" path="<directory containing this SKILL.md>/recipes/" glob="*.md" output_mode="files_with_matches"

Data

• When the user hands you a CSV, run deepline csv show --csv <path> --summary first to understand its shape (row count, columns, sample values) before deciding how to process it.
• NEVER read a large CSV into context with the Read tool. Reading CSV rows into the conversation window exhausts context and produces zero output. This is the single most common failure mode.
• Use deepline enrich for any row-by-row processing (enrichment, rewriting, research, scoring).
• To explore or understand CSV content without loading it, use deepline csv show --csv <path> --rows 0:2 for a two-row sample, or spawn an Explore subagent to answer questions about the data.
• For CSV enrichment, prefer deepline enrich --input <csv> --output <csv> --rows 0:1 ... for a one-row pilot, then rerun against the full file after inspecting output.

Tools

For signal-driven discovery (investor, funding, hiring, headcount, industry, geo, tech stack, compliance), start with deepline tools search. Do not guess fields.

Search 2-4 synonyms, execute in parallel:

deepline tools search investor
deepline tools search investor --prefix crustdata
deepline tools search --categories company_search --search_terms "structured filters,icp"
deepline tools search --categories people_search --search_terms "title filters,linkedin"

Tool search categories

Use category filters when tool type matters more than provider breadth. Common categories:

• company_search: account/company discovery tools
• people_search: people/contact discovery tools
• company_enrich: company enrichment on known companies
• people_enrich: person/contact enrichment on known people
• email_verify: email verification / deliverability
• email_finder: email lookup / discovery
• phone_finder: phone lookup / discovery
• research: company research, ad intel, job search, technographics, web research
• automation: workflow-style tools, browser/actor runs, batch automation
• outbound_tools: all Lemlist/Smartlead/Instantly/HeyReach style actions
• autocomplete: canonical filter value discovery before search
• admin: credits, monitoring, logs, schemas, local/dev utilities

Use --search_terms for extra ranking hints like structured filters, title filters, api native, autocomplete, or bulk.

Good:

• deepline tools search --categories company_search --search_terms "investors,funding"
• deepline tools search --categories research --search_terms "ads,technographics"

Avoid:

• deepline tools search stuff
• deepline tools search search across filters

2.5) Why use Deepline Enrich

When doing row by row processing (e.g. per customer, per lead, per linkedin url, etc)

Use deepline enrich as the default path.

Why:

• Row-safe: each pass is explicit and traceable.
• UI-safe: progress, errors, and outputs are visible in Session UI/Playground so your user can interject and guide you.
• Retry-safe: rerun from a known pass, not full actor chains.
• Scale-safe: large results stay in CSV lineage and are easy to inspect/filter.
• Auto-batches + rate limit safe knows how to auto batch and deal with rate limits. Almost all of the providers have rate limits that you don't know about that are managed for you if you run deepline enrich
• Lower risk: fewer custom orchestration scripts and hidden assumptions.

2.6) Session UI plan — MANDATORY for every task

Always publish your execution plan to the Session UI before running any commands. This is not optional — users monitor progress in real time via the Session UI. Without it, the UI shows nothing and users have no visibility.

# Post your plan (accepts JSON array of step labels)
deepline session start --steps '["Inspect CSV and understand shape","Search for email finder tools","Run pilot on rows 0:1","Get approval for full run","Execute full enrichment","Post-run validation and delivery"]' --user-prompt "Original user request"

# As you complete each step, update its status (0-indexed)
deepline session start --update 0 --status completed
deepline session start --update 1 --status running
deepline session start --update 1 --status completed
deepline session start --update 2 --status running
# On error:
deepline session start --update 2 --status error

Valid step statuses: pending, running, completed, error, skipped.

Live status updates within a step

As you work through a running step, send status updates to show what you're currently doing. This is for emergent work the plan couldn't predict upfront (parsing responses, falling back to alternative providers, extracting data, etc.).

# While a step is running, send status updates (attaches to the currently-running step)
deepline session status --message "Extracting company domains from provider response"
deepline session status --message "LeadMagic returned no results — falling back to ZeroBounce"
deepline session status --message "Validating 23 catch-all emails"

# Optionally target a specific step by index
deepline session status --message "Retrying with different params" --step-index 2

Each new status message marks the previous one as done and appears as the active sub-step. These are lightweight — use them freely whenever you're doing something the user would want to see.

Rules:

• Post the plan before running any enrichment/tool commands. This is step zero of every task.
• When you know the user's original request, include it on the initial deepline session start call with --user-prompt "...".
• Immediately set the first step to running right after posting the plan: deepline session start --update 0 --status running.
• Update steps as you go — mark running when starting, completed or error when done.
• Send session status messages during step execution to show what you're currently working on.
• Keep step labels short and descriptive (what, not how).
• Do not call deepline session start --steps ... at the end just to mark completion. --steps is a full set_plan replace and can wipe incremental step/sub-step history.
• Finish by updating existing steps incrementally with --update (for example, set final running step to completed).
• If --update fails with step_index ... not found (0 steps), recover by posting --steps once, then resume --update calls.
• Only re-post --steps mid-run when the plan structure truly changes.
• When writing output CSVs outside of deepline enrich, register them: deepline session output --csv <path> --label "Label".
• Use deepline session usage [--session-id UUID] [--json] when you need to inspect the current session's credits used, estimated spend, or limit state.

3) Core policy defaults

3.1 Definitions and defaults

GTM time windows, thresholds, and interpretation rules are defined in the Definitions section of enriching-and-researching.md.

Provider Playbooks

Provider-specific playbooks are bundled as separate reference files. Open the relevant playbook when provider-specific behavior, pricing, caveats, or payload conventions matter.

adyntel, ai_ark, allegrow, apify, attio, aviato, bettercontact, bloomberry, builtwith, cloudflare, contactout, crustdata, crustdata-v2, dataforseo, datagma, deepline_native, deeplineagent, discolike, dropleads, emailbison, enformion, exa, findymail, firecrawl, forager, fullenrich, generic_http, google_ads_audiences, heyreach, hubspot, hunter, icypeas, instantly, ipqs, leadmagic, lemlist, limadata, linkedin_ads_audiences, lusha, meta_audiences, openmart, opensosdata, openwebninja, parallel, peopledatalabs, predictleads, prospeo, salesforce, scrapecreators, serper, smartlead, snowflake, tamradar, theirstack, trestle, upcell, wiza, wizleads, zerobounce

• Apply defaults when user input is absent.
• User-specified values always override defaults.
• In approval messages, list active defaults as assumptions.

3.2 Working directory — set up BEFORE any file writes

NEVER write files to `/tmp/` or any absolute temp directory. Files in system /tmp/ are wiped on reboot — users permanently lose enriched CSVs, research outputs, and hours of paid enrichment work. This is a critical data-loss risk.

Set up a descriptive project-local working directory as your first action:

WORKDIR="deepline/data/<descriptive-task-slug>" && mkdir -p "$WORKDIR" && echo "$WORKDIR"

The slug must describe the task (e.g. deepline/data/yc-cmo-outbound, deepline/data/acme-email-waterfall). Do NOT use random names like mktemp generates — the user needs to find these files later. See enriching-and-researching.md for full details.

3.3 Output policy and User Interaction Pattern

• Always use deepline enrich for list enrichment or discovery at scale (>5 rows). It auto-opens a visual playground sheet so user can inspect rows, re-run blocks, and iterate.
• Even for company → ICP person flows, enrich works: search and filter as part of the process, with providers like Apify to guide.
• Even when you don't have a CSV, create one and use deepline enrich.
• This process requires iteration; one-shotting via deepline tools execute is short sighted.
• For run_javascript in deepline enrich, put JS in payload.code; the current row is auto-injected as row at runtime, so you usually should not pass row yourself.
• If a command created CSV outside enrich, register it with the Session UI so a table card appears: deepline session output --csv <csv_path> --label "My Results". This is the lightweight alternative to deepline enrich for surfacing output in the Session UI.
• When execution work is complete, stop backend explicitly with deepline backend stop --just-backend unless the user asked to keep it running.
• In chat, send the file path + playground status, not pasted CSV rows, unless explicitly requested.
• Preserve lineage columns (especially _metadata) end-to-end. When rebuilding intermediate CSVs with shell tools, carry forward _metadata columns.
• Never enrich a user-provided or source CSV in-place. Use --output to write to your working directory on the first pass, then --in-place on that output for subsequent passes. --in-place is for iterating on your own prior outputs — never on source files.
• For reruns, keep successful existing cells by default; use --with-force <alias> only for targeted recompute.

See enriching-and-researching.md for deepline csv commands, pre-flight/post-run script templates, and inspection details.

3.4 Final file + playground check (light)

• Keep one intended final CSV path: FINAL_CSV="${OUTPUT_DIR:-$WORKDIR}/<requested_filename>.csv"
• Before finishing: use the post-run inspection script pattern from enriching-and-researching.md. Run it once instead of separate checks.
• In the final message, always report: exact FINAL_CSV and exact Playground URL.
• Before closing the session, follow the Section 7 consent step for session sharing.

4) Credit and approval gate (paid actions)

4.1 Required run order

1. Pilot on a narrow scope (example --rows 0:1 for one row).
2. Request explicit approval.
3. Run full scope only after approval.

4.2 Execution sizing

• Use smaller sequential commands first.
• Keep limits low and windows bounded before scaling.
• For TAM sizing, a great hack is to keep limits at 1 and most providers will return # of total possible matches but you only get charged for 1.
• Prefer providers and plays that charge on returned results or successful hits when coverage is uncertain. If a provider bills per attempt/request/page, prove quality on a tiny pilot before letting it fan out.
• Stop after the pilot when the first rows show low usable coverage, wrong-person/company matches, missing getters, or high cost per usable row. Change route/provider order before buying the same failure at full scale.
• Do not depend on monthly caps as a hard risk control.

4.2.1 Over-provision, then filter — never chase missing rows

When the user asks for N rows, start with ~1.4×N (e.g., 35 for 25). Every pipeline phase has natural falloff — contact search misses ~15-20% of companies, email waterfall misses ~5-10% of contacts. Fighting to complete the hard rows is almost always a waste: the companies that providers can't find contacts for are the same ones that won't have email coverage either.

Do this:

1. Pull more candidates than needed at the top of funnel.
2. Run the full pipeline (contacts → emails → outbound).
3. At the end, filter to the best N complete rows and deliver those.
4. Drop incomplete rows — don't retry or manually patch them.

Do NOT do this:

• Trim results to exactly N before running the pipeline.
• Spend turns retrying failed lookups with fallback providers, deeplineagent research passes, or manual patching.
• Run enrichment on all rows just to fill gaps in a few (especially broad deeplineagent research passes).

Provider coverage is a property of the company, not something you can overcome with more effort. Tiny startups with 5 people will have zero coverage across all providers — no amount of retrying changes that. Over-provision at the top and let incomplete rows fall off naturally.

4.3 Approval message content

Include all of:

1. Provider(s)
2. Pilot summary and observed behavior
3. Intent-level assumptions (3–5 one-line bullets)
4. CSV preview from a real deepline enrich --rows 0:1 one-row pilot
5. Credits estimate / range
6. Full-run scope size
7. Max spend cap
8. Approval question: Approve full run?

Note: deepline enrich already prints the ASCII preview by default, so use that output directly.

Strict format contract (blocking):

1. Use the exact four section headers: Assumptions, CSV Preview (ASCII), Credits + Scope + Cap, Approval Question.
2. If any required section is missing, remain in AWAIT_APPROVAL and do not run paid/cost-unknown actions.
3. Only transition to FULL_RUN after an explicit user confirmation to the approval question.
4. run_javascript is the non-AI path. aiinference is for general classification/structured reasoning, and deeplineagent is for context gathering / web research / signal extraction.

Approval template:

Assumptions

- <intent assumption 1>
- <intent assumption 2>

CSV Preview (ASCII)
<paste verbatim output from deepline enrich --rows 0:1>
Credits + Scope + Cap

- Provider: <name>
- Estimated credits: <value or range>
- Full-run scope: <rows/items>
- Spend cap: <cap>
- Pilot summary: <one short paragraph>

Approval Question
Approve full run?

4.4 Mandatory checkpoint

• Must run a real pilot on the exact CSV for full run (--rows 0:1, end exclusive).
• Must include ASCII preview verbatim in approval.
• If pilot fails, fix and re-run until successful before asking for approval.
• Before using AskUserQuestion for the approval gate, notify the Session UI so the user knows to check the terminal:

  deepline session alert --message "Approval needed: run enrichment on N rows (~X credits)"

4.5 Billing commands

deepline billing balance  # Show current credit balance
deepline billing usage    # Show recent billing activity and grouped recent usage
deepline billing limit    # Show the current monthly billing cap

When credits at zero, link to https://code.deepline.com/dashboard/billing to top up. 10 credits == 1 USD

5) Provider routing (high level)

Reminder: you should have already read the relevant sub-doc from Section 2 before reaching this point. If you haven't, go back and read it now. This section is a quick-reference summary, NOT a substitute for the sub-docs.

• Search / discovery → You MUST have finding-companies-and-contacts.md open. It contains the parallel execution patterns, provider filter schemas, and provider mix tables. Start with deepline tools search <intent> and execute field-matched provider calls in parallel; when the deepline-list-builder subagent is available, use subagent-based parallel search orchestration as the preferred pattern. Use deeplineagent only for synthesis or ambiguity resolution after the direct discovery path is exhausted.
• Enrich / waterfall / coalesce → You MUST have enriching-and-researching.md open. It contains deepline enrich syntax, play routing guidance, waterfall column patterns, and coalescing logic. Do not restate play internals from memory; treat the play itself as the source of truth for exact provider order and gating.
• Custom signals / messaging → Read enriching-and-researching.md (custom signals section). Use run_javascript for deterministic transforms/template logic and deeplineagent for AI work. Start from prompts.json.
• Verification → leadmagic_email_validation first, then enrich corroboration.
• LinkedIn scraping -> Apify actors, by far the best. Use deepline tools describe apify_run_actor_sync to see the available actors or search for more.
• For phone recovery, read enriching-and-researching.md and follow the notes/provider guidance there rather than relying on deleted numbered sections.

Provider path heuristics:

• Broad first pass: direct tool calls for high-volume discovery.
• Quality pass: AI-column orchestration with explicit retrieval instructions.
• For job-change recovery: prefer quality-first (crustdata_person_enrichment, peopledatalabs_*) before leadmagic_* fallbacks.
• Never treat one provider response as single-source truth for high-value outreach.

6) Additional notes

Critical: keep writing-outreach.md workflow context active when running any sequence task. It is not optional for ICP-driven messaging.

Apify actor flow (short canonical policy)

Operational troubleshooting: rate limits and CLI health

• Use deepline enrich for heavy row-by-row work whenever possible. It has built-in rate-limit handling (adaptive retries/backoff) for standard upstream limits. If you are building a homegrown script, assume it does not include the same automatic protection unless you explicitly implement it.
• If enrichment or CLI behavior is unstable, rerun the installer to ensure the latest CLI/client wiring is in place:

curl -s "https://code.deepline.com/api/v2/cli/install" | bash

Sites requiring auth: Don't use Apify. Tell the user to use Claude in Chrome or guide them through Inspect Element to get a curl command with headers (user is non-technical).

1. If user provides actor ID/name/URL: use it directly.
2. If not, search deepline tools describe apify_run_actor_sync for the actor id, or try deepline tools search.
3. If not present, run discovery search.
4. Avoid rental-priced actors.
5. For LinkedIn post scraping, prefer supreme_coder/linkedin-post for generic posts/search URLs and harvestapi/linkedin-post-reactions when the goal is engagers/reactions. Avoid silentflow/linkedin-posts-scraper-ppr and alizarin_refrigerator-owner/linkedin-post-scraper unless the user explicitly asks for them.
6. Pick high rating plus high usage/run count; when tied, choose best evidence-quality/price balance.
7. Honor operatorNotes over public ratings when conflicting.

deepline tools execute apify_list_store_actors --payload '{"search":"linkedin company employees scraper","sortBy":"relevance","limit":20}'
deepline tools execute apify_get_actor_input_schema --payload '{"actorId":"bebity/linkedin-jobs-scraper"}'

7) Feedback & session sharing

7.1 Proactive issue reporting (mandatory)

Do not wait for the user to ask. If there is a meaningful failure, send feedback proactively using deepline feedback send.

Trigger when any of these happen:

• A provider/tool call fails repeatedly.
• Output is clearly wrong for the requested task.
• A CLI/runtime bug blocks completion.
• You needed a significant workaround to finish.

Run once per issue cluster (avoid spam), and include:

• workflow goal
• tool/provider/model used
• failure point and exact error details
• reproduction steps attempted

deepline feedback send "Goal: <goal>. Tool/provider/model: <details>. Failure: <what broke>. Error: <exact message>. Repro attempted: <steps>."

7.2 End-of-session consent gate (mandatory)

At the end of every completed run/session, ask exactly one Yes/No question:

Would you like me to send this session activity to the Deepline team so they can improve the experience? (Yes/No)

If user says:

• Yes -> run:

  deepline sessions send --current-session

• No -> do not send the session.

Ask once per completed run. Do not nag or re-ask unless the user starts a new run/session.