Skip to main content

COS MCP Server

MCP server for the Communications Optimization System (COS) — analyze content for engagement, personality fit, strategic clarity, and cognitive framing.

Quick Start

# Install
pip install cos-mcp

# Set your API key
export COS_API_KEY="your-cos-api-key"

# Run (stdio mode for Claude Desktop / Claude Code)
cos-mcp

Claude Code

claude mcp add cos-mcp -- cos-mcp

Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "cos": {
      "command": "cos-mcp",
      "env": {
        "COS_API_KEY": "your-cos-api-key"
      }
    }
  }
}

Remote (zero-install)

claude mcp add --transport http cos https://mcp.semalytics.io/mcp \
  --header "Authorization: Bearer YOUR_COS_API_KEY"

Tools

Tool Credits Description
chat 1 Conversational COS analysis agent
audience_profile 1 OCEAN profile + Cialdini principle from an audience description
analyze_framework 2 Single framework (HAPE, Big Five, Strategic Clarity, Framing)
analyze_platform 2 Platform-specific optimization (13 platforms)
analyze_quality 2 5-dimension quality assessment
execute_template 2 Run pre-built analysis templates
analyze_persuasion 3 Domain-specific persuasion analysis
profile_agent 3 Infer personality from writing samples
analyze_content 5 Full 4-framework analysis
optimize_email_for_prospect 8 CRM composite — audience profile + draft generation + persuasion + platform scoring in one call
analyze_full_comms 10 All 7 frameworks in parallel
get_templates 0 List available templates
get_template_details 0 Get template details

CRM integration — optimize_email_for_prospect

Built for Sales CRMs (Clarify, HubSpot, Salesforce, Pipedrive). Pass a Person/Account context, get back a draft + full scoring in a single call.

optimize_email_for_prospect(
  audience_description="VP Eng at 200-person SaaS, viewed pricing 3x this week",
  intent="cold_outreach",
  name="Jamie Lee",
  title="VP Engineering",
  company="Acme Co",
  recent_signals=["pricing-page visits", "engineering team hiring"],
  sender_context="We cut CI costs by 40% for engineering teams.",
  # draft=None → COS generates one. Pass a draft to score+refine instead.
)

Returns: {audience_profile, agent_profile?, draft, draft_was_generated, persuasion, platform, rewrites[], one_thing, cialdini_principle}.

Get an API Key

Sign up at semalytics.com/cos to get your API key and free starter credits.

License

MIT

Metadata

Release files for cos-mcp 0.3.1

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for cos-mcp 0.3.1
File Size Uploaded Provenance Actions
cos_mcp-0.3.1.tar.gz 11.2 kB Without provenance Details for cos_mcp-0.3.1.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for cos-mcp 0.3.1
File Interpreter ABI Platform Actions
cos_mcp-0.3.1-py3-none-any.whl Python 3 none any Details for cos_mcp-0.3.1-py3-none-any.whl

Total release size: 22.6 kB

Release files / cos_mcp-0.3.1.tar.gz

Download URL cos_mcp-0.3.1.tar.gz
Size 11.2 kB
Tags Source
SHA-256 checksum
How to use checksums
c95f1f1e90349c89fe7b4e2c46f7262aaafaefccf376870a89a082004f17e728
BLAKE2b-256 checksum
How to use checksums
182340d73a1205a9f8fbb837ec9e8d674b540581309fbbd9de5693ce3fd12939
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.11.13

Release files / cos_mcp-0.3.1-py3-none-any.whl

Download URL cos_mcp-0.3.1-py3-none-any.whl
Size 11.5 kB
Tags Python 3
SHA-256 checksum
How to use checksums
e1838b8cfcba825bfe523ebdccc3572273f4a89fa22e7c4c236f819440f1aaef
BLAKE2b-256 checksum
How to use checksums
8adeca010f201b8659f5de844b888aeb4b2d2ed5e9fd2e237549c5b76674f6d1
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.11.13

Security for cos-mcp 0.3.1

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page