Cited CLI

Command-line interface for the Cited GEO (Generative Engine Optimization) platform. Audit how often your business is cited by AI assistants, generate recommendations, and produce AI-ready content solutions — all from the terminal or a Python script.

Installation

Homebrew (recommended)

brew tap youcited/cited
brew install cited

From source (development)

git clone https://github.com/YouCited/cited-cli.git
cd cited-cli
pip install -e ".[dev]"

Quick Start

# Log in (opens browser by default)
cited login

# Check you're connected
cited status

# List your businesses
cited business list

# See all commands
cited --help
cited audit --help
cited audit template --help

For the full end-to-end workflow — creating a business, running a GEO audit, and generating content solutions — see the GEO Audit Pipeline section below.

---

Authentication

# Browser login (default — opens youcited.com login page)
cited login

# Specific OAuth provider
cited login --provider google
cited login --provider microsoft
cited login --provider github

# Password login (non-interactive / CI)
cited login --email you@example.com --password "Secret123!"

# Check status
cited auth status

# Log out
cited logout

Password registration is a two-step email-verification flow:

cited register --email you@example.com --name "Your Name" --password "Secret123!"
# → Check your inbox, paste the verification URL when prompted

---

MCP Server

This repo also includes cited-mcp, a Model Context Protocol server that exposes 56 tools for AI assistants like Claude. It lets Claude manage businesses, run GEO audits, generate recommendations, work a prioritized action plan, verify fixes via the Validation Engine, and create solutions on your behalf.

Claude Desktop (Custom Connector — recommended)

No local install required. Claude connects to Cited's hosted MCP server directly:

1. Open Claude Desktop → Settings → Customize > Connectors
2. Click "+" → "Add custom connector"
3. Enter URL: https://mcp.youcited.com/mcp
4. Click Add

Your browser will open for authentication on first use. See Anthropic's connector guide for details.

Requires: Claude Pro, Max, Team, or Enterprise plan.

Claude Desktop (Developer Config — alternative)

If Custom Connectors aren't available, use the developer config with mcp-remote:

{
  "mcpServers": {
    "cited": {
      "command": "npx",
      "args": ["mcp-remote", "https://mcp.youcited.com/mcp"]
    }
  }
}

Add to Settings → Developer → Edit Config, then restart Claude Desktop.

Prerequisite: Node.js (LTS) must be installed for npx.

Claude Desktop (local stdio)

{
  "mcpServers": {
    "cited": {
      "command": "uvx",
      "args": ["cited-mcp"],
      "env": {
        "CITED_TOKEN": "your-jwt-token"
      }
    }
  }
}

Get your token with cited login then cited auth token.

Cursor

Add a .cursor/mcp.json file to your project root (or ~/.cursor/mcp.json for global access):

{
  "mcpServers": {
    "cited": {
      "command": "npx",
      "args": ["mcp-remote", "https://mcp.youcited.com/mcp"]
    }
  }
}

Verify Node.js is installed (node --version), or install with brew install node.

Open Settings (Cmd+,) → search "MCP" → verify the server shows a green status. Use tools in Composer (Agent mode).

Updating to new Cited features

Cited adds new tools regularly. To access new tools after a release, disconnect and reconnect the Cited connector in your Claude Desktop or Claude.ai connector settings — restarting Claude alone is not enough, because the connector's tool list is cached at registration time. Once reconnected, ask the agent to call whats_new to see what changed since you last connected.

Example

Once connected, ask Claude something like:

"List my businesses, then run a GEO audit on Acme Corp using the default template. When it's done, generate recommendations and show me the top insights."

Claude will chain the tools automatically — list_businesses → start_audit → get_audit_status → start_recommendation → get_recommendation_insights.

Available Tools

Category	Tools
Auth	check_auth_status, login, logout
Businesses	list_businesses, get_business, create_business, update_business, delete_business, crawl_business, get_health_scores, get_usage_stats
Audit Templates	list_audit_templates, get_audit_template, create_audit_template, update_audit_template, delete_audit_template
Audits	start_audit, get_audit_status, get_audit_result, list_audits, export_audit
Recommendations	start_recommendation, get_recommendation_status, get_recommendation_result, get_recommendation_insights, list_recommendations
Solutions	start_solution, get_solution_status, get_solution_result, list_solutions
Jobs	get_job_status, cancel_job
HQ	get_business_hq
Analytics	get_analytics_trends, get_analytics_summary, compare_audits
Agent API	get_business_facts, get_business_claims, get_competitive_comparison, get_semantic_health, buyer_fit_query
Action Plan	get_action_plan, get_quick_wins, mark_action_done, dismiss_action, get_action_progress

See packages/mcp/README.md for full details, environment variables, and development instructions.

---

Global Flags

These flags apply to every command and must be placed immediately after cited:

Flag	Short	Description
--json	-j	Machine-readable JSON output (great for scripting)
--text	-t	Human-readable text output (overrides output config)
--env	-e	Environment: dev, prod, local (default: prod)
--profile	-p	Config profile to use
--verbose	-v	Debug logging
--quiet	-q	Minimal output
--no-color		Disable colors (auto-applied when stdout is not a TTY)

---

GEO Audit Pipeline

The complete workflow from business registration through AI-ready content solutions. Each step captures an ID for the next — just like piping commands together.

Prerequisites

• jq installed (brew install jq) for ID extraction in bash
• A publicly resolvable website domain (not example.com)
• Account logged in: cited login

Step 1 — Create your business profile

BUSINESS_ID=$(cited --json business create \
  --name        "Acme Corp" \
  --website     "https://acme.com" \
  --description "Acme builds AI-powered tools for enterprise productivity teams." \
  --industry    "technology" \
  | jq -r '.id')

echo "Business: $BUSINESS_ID"

--industry must be one of: automotive beauty consulting education entertainment finance fitness government healthcare home_services hospitality legal manufacturing non_profit real_estate restaurant retail technology other

Step 2 — Scan your website

The crawler reads your site so the platform understands your content, products, and brand signals.

CRAWL_JOB=$(cited --json business crawl $BUSINESS_ID | jq -r '.job_id')
cited job watch $CRAWL_JOB

# View crawl results as health score bars
cited business health $BUSINESS_ID

Step 3 — Create an audit template

An audit template defines the questions you want AI assistants to answer. Ask questions your target buyers would actually type.

TEMPLATE_ID=$(cited --json audit template create \
  --name     "Q4 GEO Audit" \
  --business $BUSINESS_ID \
  --question "Are we cited when people ask about AI productivity tools?" \
  --question "Does our product appear in enterprise software recommendations?" \
  --question "Are we mentioned when users research our core use case?" \
  | jq -r '.id')

echo "Template: $TEMPLATE_ID"

# List all templates for a business
cited audit template list --business $BUSINESS_ID

# Inspect a template
cited audit template get $TEMPLATE_ID

Auto-generated questions are rarely perfect out of the box. Refine them before running the audit:

# Replace all questions (--question flags replace the full list)
cited audit template update $TEMPLATE_ID \
  --question "Are we cited when enterprise buyers ask about AI safety?" \
  --question "Does our product appear in responsible AI recommendations?" \
  --question "Are we mentioned when developers compare AI platforms?"

# Update just the name (keeps existing questions)
cited audit template update $TEMPLATE_ID --name "Q4 GEO Audit v2"

# Verify changes
cited audit template get $TEMPLATE_ID

Step 4 — Run the audit

AUDIT_JOB=$(cited --json audit start $TEMPLATE_ID --business $BUSINESS_ID \
  | jq -r '.job_id')

# Watch live progress bar
cited job watch $AUDIT_JOB

# View results: per-question citation data across providers
cited audit result $AUDIT_JOB

Step 5 — Generate recommendations

RECO_JOB=$(cited --json recommend start $AUDIT_JOB | jq -r '.job_id')
cited job watch $RECO_JOB

Step 6 — Discover what to fix

recommend insights prints a table of all actionable items with the source IDs you'll need for the next step:

cited recommend insights $RECO_JOB

                          Available Insights
┏━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━┓
┃ # ┃ Type              ┃ Source ID                ┃ Description              ┃
┡━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━┩
│ 1 │ question_insight  │ d5b17bda-fa5b-48ff-...   │ Are we cited for AI...   │
│ 2 │ head_to_head      │ wikipedia.org            │ wikipedia.org            │
│ 3 │ strengthening_tip │ llms_txt                 │ Create llms.txt for AI   │
└───┴───────────────────┴──────────────────────────┴──────────────────────────┘

Run a solution: cited solution start $RECO_JOB --type <type> --source <source_id>

Step 7 — Generate a content solution

Pick a row from the insights table and pass its type and source ID:

cited solution start $RECO_JOB \
  --type   question_insight \
  --source d5b17bda-fa5b-48ff-8238-a644c78b404a

# → ✓ Solution Started
# → Track progress: cited job watch <sol_job_id>
# → View artifacts: https://app.youcited.com/solutions/<sol_job_id>

--type values: question_insight · head_to_head · strengthening_tip · priority_action

Rich solution artifacts (blog posts, FAQ schemas, structured content) are rendered in the web app. The CLI prints the direct link so you can jump straight there.

---

Scripting & Automation

Every command supports --json output, making the CLI a first-class building block for automation pipelines — in bash, Python, or any language that can run a subprocess.

Bash with jq

# Capture IDs by piping --json output through jq
BUSINESS_ID=$(cited --json business create --name "..." --website "..." \
  --description "..." --industry "technology" | jq -r '.id')

TEMPLATE_ID=$(cited --json audit template create \
  --name "My Audit" --business $BUSINESS_ID \
  --question "Are we cited for X?" | jq -r '.id')

AUDIT_JOB=$(cited --json audit start $TEMPLATE_ID | jq -r '.job_id')
cited job watch $AUDIT_JOB

RECO_JOB=$(cited --json recommend start $AUDIT_JOB | jq -r '.job_id')
cited job watch $RECO_JOB

# Extract first question insight source_id
SOURCE_ID=$(cited --json recommend result $RECO_JOB \
  | jq -r '.question_insights[0].question_id')

cited solution start $RECO_JOB --type question_insight --source $SOURCE_ID

Python

The CLI is a thin, authenticated HTTP wrapper — Python scripts can drive it directly via subprocess with no extra dependencies. This is the same pattern used by the integration test suite.

#!/usr/bin/env python3
"""
cited-pipeline.py — Automate a full GEO audit pipeline.

Requirements: `cited` CLI installed and authenticated (`cited login` first).
Usage:        python3 cited-pipeline.py
"""
import json
import subprocess

ENV = "dev"  # or "prod"


def cited_json(*args: str) -> dict | list:
    """Run a cited command with --json and return parsed output."""
    result = subprocess.run(
        ["cited", "--env", ENV, "--json", *args],
        capture_output=True, text=True, check=True,
    )
    return json.loads(result.stdout)


def cited_watch(*args: str) -> None:
    """Run a cited command streaming human output live (progress bars, tables)."""
    subprocess.run(["cited", "--env", ENV, *args], check=True)


# ── Step 1: Create business ───────────────────────────────────────────────────
print("→ Creating business...")
business = cited_json(
    "business", "create",
    "--name",        "Acme Corp",
    "--website",     "https://acme.com",
    "--description", "Acme builds AI-powered tools for enterprise productivity teams.",
    "--industry",    "technology",
)
business_id = business["id"]
print(f"  Business: {business_id}")

# ── Step 2: Scan the website ──────────────────────────────────────────────────
print("→ Scanning website...")
crawl = cited_json("business", "crawl", business_id)
if crawl_job := crawl.get("job_id"):
    cited_watch("job", "watch", crawl_job)

# ── Step 3: Create audit template ────────────────────────────────────────────
print("→ Creating audit template...")
template = cited_json(
    "audit", "template", "create",
    "--name",     "Q4 GEO Audit",
    "--business", business_id,
    "--question", "Are we cited when people ask about AI productivity tools?",
    "--question", "Does our product appear in enterprise software recommendations?",
    "--question", "Are we mentioned when users research our core use case?",
)
template_id = template["id"]
print(f"  Template: {template_id}")

# ── Step 3b: Refine questions (auto-generated questions usually need tuning) ─
print("→ Refining template questions...")
updated = cited_json(
    "audit", "template", "update", template_id,
    "--question", "Are we cited when enterprise buyers ask about AI safety?",
    "--question", "Does our product appear in responsible AI recommendations?",
    "--question", "Are we mentioned when developers compare AI platforms?",
)
print(f"  Updated: {len(updated['questions'])} questions")

# ── Step 4: Run the audit ─────────────────────────────────────────────────────
print("→ Starting audit...")
audit = cited_json("audit", "start", template_id, "--business", business_id)
audit_job = audit["job_id"]
cited_watch("job", "watch", audit_job)

# ── Step 5: Generate recommendations ─────────────────────────────────────────
print("→ Generating recommendations...")
recommend = cited_json("recommend", "start", audit_job)
reco_job = recommend["job_id"]
cited_watch("job", "watch", reco_job)

# ── Step 6: View available insights ──────────────────────────────────────────
print("→ Available insights:")
cited_watch("recommend", "insights", reco_job)

# ── Step 7: Start a content solution ─────────────────────────────────────────
# Extract source IDs directly from the JSON result
result = cited_json("recommend", "result", reco_job)
question_insights = result.get("question_insights", [])

if question_insights:
    source_id = question_insights[0]["question_id"]
    print(f"→ Solving: {question_insights[0]['question_text'][:60]}...")
    solution = cited_json(
        "solution", "start", reco_job,
        "--type",   "question_insight",
        "--source", source_id,
    )
    sol_job = solution["job_id"]
    print(f"  Solution job:    {sol_job}")
    print(f"  View artifacts:  https://{ENV}.youcited.com/solutions/{sol_job}")

print("\n✓ Pipeline complete!")

The cited_json() / cited_watch() pattern — capture IDs from --json output, stream human output for long-running jobs — is exactly what the integration test suite uses internally with mocked HTTP responses.

---

Command Reference

cited login                               Log in (top-level alias)
cited logout                              Log out (top-level alias)
cited register                            Register new account (top-level alias)
cited status                              API health check
cited version                             Show CLI version

cited auth login|logout|register|status|token

cited config set|get|show|environments

cited business list|get|create|update|delete|health|crawl

cited audit template list|get|create|update|delete   ← Manage audit templates
cited audit start|status|result|list|export   ← Run and inspect audits

cited recommend start|status|result|list      ← Generate recommendations
cited recommend insights                      ← View actionable source table

cited solution start|status|result|list       ← Generate content solutions

cited hq <business_id> [--full] [--personas] [--products] [--intents] [--actions]

cited analytics compare|trends|summary

cited agent facts|claims|comparison|semantic-health|buyer-fit

cited job watch|cancel

---

Configuration

Config is stored in ~/.cited/config.toml:

# Set default environment (saves typing --env dev every time)
cited config set environment dev

# Set default output format (saves typing --json every time)
cited config set output json    # or "text" (default)

# Set a default business (used by commands that accept --business)
cited config set default_business_id <uuid>

# Set Agent API key (for cited agent commands)
cited config set agent_api_key <key>

# Inspect current config
cited config show
cited config environments

Tip: With output set to json, use --text on any command to get human-readable output for that one invocation.

---

JSON Output

All commands write clean JSON to stdout with --json, enabling pipelines with jq, python3 -c, or any other tool:

# Extract specific fields
cited --json business list | jq '.[].name'
cited --json audit result <id> | jq '.overall_citation_rate'
cited --json recommend result <id> | jq '.question_insights[].question_text'

# Save full results
cited --json audit result <id> > audit.json
cited --json recommend result <id> > recommendations.json

# Count citations by provider
cited --json audit result <id> | jq '[.citations_pulled[].provider_name] | group_by(.) | map({provider: .[0], count: length})'

Errors always go to stderr as {"error": true, "message": "..."} so they never corrupt stdout pipelines.

---

Exit Codes

Code	Meaning
0	Success
1	General error
2	Authentication error
3	Not found
4	Validation error
5	Rate limited

---

Development

pip install -e ".[dev]"
pytest -v                                       # All tests
pytest tests/test_pipeline.py -v               # Pipeline integration tests
ruff check src/
mypy src/cited_cli/ --ignore-missing-imports

MCP test harness

scripts/mcp_test_harness.py runs an end-to-end test suite against the deployed MCP server (dev or prod). Exercises the four-PR-plan test cases — tools_fingerprint on ping, whats_new lookups, upgrade_plan no-op response, tool_unavailable error shape — plus auth/host-header/rate-limit/cross-env/robots.txt regression checks.

./scripts/mcp_test_harness.py --env dev    # mcpdev.youcited.com
./scripts/mcp_test_harness.py --env prod   # mcp.youcited.com

First run per env opens a browser for OAuth; tokens are cached at /tmp/cited-mcp-{env}-token.json and reused for 7 days. The rate-limiter test fills the user's per-token window for ~60s, so run it last in any debug session.

Releasing

Release and deployment scripts are in a private infrastructure repo. See that repo's README for release and deploy instructions.