Pytch MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| PYTCH_API_KEY | Yes | Your Pytch API key, starting with pytch_live_... | |
| PYTCH_API_URL | No | The Pytch API base URL | https://api.pytch.cc |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| check_domain_deliverabilityA | Check domain email deliverability (F6 Block): Performs live DNS audit for SPF, DMARC, DKIM, MX, and DNSBL spam blacklists. Calculates a 0-100 Deliverability Health Score and generates a safe ramp-up warmup schedule. Args: domain: Domain name or email address (e.g. 'company.com' or 'ceo@company.com') selector: Optional DKIM selector if known (e.g. 'google', 'selector1') |
| check_ai_brand_visibilityA | Audit brand visibility in AI answers (F9 Block / GEO): Simulates searches across LLMs (ChatGPT/Perplexity/Gemini style), evaluates whether the brand is cited, calculates AI Share of Voice (SOV), extracts media outlets influencing AI responses, and matches them to journalists. Args: brand_name: Name of the company or brand (e.g. 'Pytch') category: Industry niche or product category (e.g. 'AI PR agent', 'fintech') website: Optional brand website URL competitors: Optional list of competitor brand names |
| search_journalistsA | Search verified tech and business journalists by beat, topic or media publication. Args: topic: Keyword or beat (e.g. 'AI', 'B2B SaaS', 'Biotech', 'Venture Capital') outlet: Optional media publication filter (e.g. 'TechCrunch', 'Forbes', 'VentureBeat') limit: Max contacts to return (default: 10) |
| list_pr_projectsA | List active PR campaigns and projects for the authenticated user. Requires PYTCH_API_KEY environment variable. |
| get_project_resultsB | Get identified media contacts, pitch status, and response activity for a project. Args: project_id: UUID of the project in Pytch limit: Number of contacts to fetch |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 5 tools
Each tool targets a distinct resource and action: two self-contained audit blocks (domain deliverability, AI brand visibility), a journalist directory search, and two project-scoped read operations. There is no plausible confusion between email deliverability auditing and AI Share of Voice auditing, nor between listing projects and fetching one project's results.
All five tools follow a clean snake_case verb_noun pattern (check_*, search_*, list_*, get_*), with the verb accurately reflecting the operation's read/analysis semantics. No mixed casing or inconsistent verb styles.
Five tools is on the lean side but each maps to a coherent capability rather than being filler, so nothing feels redundant. It is slightly thin for a PR platform surface, since the audit blocks and the project reads are not joined by any action tool.
The surface covers discovery (search_journalists), monitoring (list_pr_projects, get_project_results) and two standalone audits, but the core PR lifecycle is severed: there is no way to create a project, add/pitch a journalist, or log outreach status. Agents can read state and audit domains but cannot perform the primary action the product implies.