Skip to main content
Use this when you want an agent to run GTM work without opening the Deepline UI.

​
Pick the integration path

Agent environmentBest pathNotes
Claude CodeDeepline skill + CLIBest covered today. See Claude Code.
CodexDeepline skill + CLIInstall the hosted skills, then run commands from the agent terminal.
CursorDeepline skill + CLIInstall hosted skills, or point Cursor at the skill markdown files.
Gemini CLIDeepline skill + CLISupported by the hosted skill installer.
AntigravityDeepline skill + CLISupported by the hosted skill installer.
Hermes Agent / Hermes IDECLIUse the Deepline CLI from the shell. Deepline does not publish a hosted Hermes skill target yet.
Any other shell-capable agentCLIInstall, authenticate, then give the agent the exact command to run.
The rule is simple: if the agent can run shell commands, it can run Deepline. Use skills where the agent supports them. Use the plain CLI everywhere else.

​
Install Deepline once

curl -fsSL "https://code.deepline.com/api/v2/cli/install" | bash
deepline auth register --wait auto
deepline auth status
If the agent runs inside a sandbox, make sure its shell can see the installed deepline binary before asking it to enrich a CSV. 
deepline -h

​
Install hosted skills where supported

npx skills add "https://code.deepline.com/.well-known/skills/index.json" --agents codex claude-code cursor gemini-cli antigravity --global --yes --skill '*' --full-depth
This installs the same Deepline skill bundle for the supported agent targets. The hosted installer does not currently include Hermes as a target. For Hermes, use the CLI path above and link the agent to this page or to the GTM agent repo.

​
Give the agent a concrete job

Good agent instructions name the file, output file, run name, row range, and tool payload. 
deepline enrich --input leads.csv --output enriched.csv --name "email-pilot" --rows 0:4 \
  --with '{"alias":"email","tool":"name_and_domain_to_email_waterfall","payload":{"first_name":"{{first_name}}","last_name":"{{last_name}}","domain":"{{domain}}"}}'
If the pilot looks right, run the same command without the row range: 
deepline enrich --input leads.csv --output enriched.csv --name "email-full" \
  --with '{"alias":"email","tool":"name_and_domain_to_email_waterfall","payload":{"first_name":"{{first_name}}","last_name":"{{last_name}}","domain":"{{domain}}"}}'
Do not ask the agent to guess flags. The canonical command shape is: 
deepline enrich --input <file.csv> --output <file.csv> --name <run-name> --with <json> [--rows start:end]

​
Use the GTM agent repo

For repo-backed agent workflows, start from the Deepline GTM agent project:

deepline-gtm-agent

Reference repo for improving Deepline GTM agent workflows across coding agents.
Use that repo when you want versioned prompts, reusable workflows, or a shared agent setup that works across Codex, Claude Code, and other shell-capable coding agents.

​
Troubleshooting

Open the agent terminal and run deepline -h. If that fails, reinstall the CLI in the same sandbox or add the install prefix to the agent shell’s PATH.
Use only the supported hosted targets: codex, claude-code, cursor, gemini-cli, and antigravity. For Hermes or another agent, use the CLI directly.
Stop and check Content canonicals before running it. For enrichment, use the command shape on this page.