Craft Gtm Mcp — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited Craft Gtm Mcp (Agent Skill) and scored it 65/100 (yellow). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 3 high-severity and 2 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 5 flagged
The text {match} is the classic direct prompt-injection phrasing. Placed in a skill body that the agent reads as trusted instructions, it tries to make the agent abandon its prior rules and follow whatever comes next — a full system-prompt override.
ignore/disregard/forget … previous instructions sentence.The text {match} asks the agent to disclose its hidden system prompt or initial instructions. That is often the first step of a larger attack: knowing the system prompt lets an attacker craft inputs that defeat its constraints by mimicking its own voice.
repeat/reveal/print your system prompt request from the skill.Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
CRAFT GTM Framework MCP Server: a complete redesign with metric parsing, context-aware outputs, and a structured draft with placeholders where your inputs give no fact.
Add https://craft-gtm.gtmhelix.com/mcp to Claude or ChatGPT as a custom connector. It needs no sign-in and always runs the newest version (2.2.23). The same tools run as a free web app with a form per tool at https://craft-gtm.gtmhelix.com/, and the setup steps are at https://craft-gtm.gtmhelix.com/connect/.
The npm package below is an older version (2.0.1 on npm on 27 September 2026) until the next npm release. Use it only if you need a local stdio server.
This is a major redesign addressing all critical issues from v1.x:
| Issue | v1.x | v2.0.0 |
|---|---|---|
| PMF Scorecard | Blank ___ outputs | Actually parses and scores your metrics |
| Launch Commander | Same 12-week template for all | Adapts to launch type, team size, budget |
| Retention Playbook | Generic health scores | Parses churn reasons, creates specific interventions |
| Competitive Intel | [Research] placeholders | Uses your competitor data, generates battle cards |
| CRAFT Analyzer | Blank scorecard output | Actually analyzes documents, finds gaps |
npm install -g @shashwatgtmalpha/craft-gtm-mcpOr add to your Claude Desktop config:
{
"mcpServers": {
"craft-gtm": {
"command": "npx",
"args": ["-y", "@shashwatgtmalpha/craft-gtm-mcp"]
}
}
}Generated on 27 September 2026 from the server's own tool list and checked again on 17 October 2026 against tools/list of craft-gtm-mcp 2.2.23 (the same code as the hosted MCP address), so every tool name, title, description and input below is exactly what the server accepts. Every tool is read-only.
| # | Tool | Title | What it does |
|---|---|---|---|
| 1 | pmf_scorecard | PMF Scorecard | Generate a Product-Market Fit scorecard. Parses the metrics you provide (MRR, churn, NPS, CAC, LTV, retention, activation and similar) and scores each dimension against built-in benchmark ranges. |
| 2 | launch_commander | Launch Commander | Generate a context-aware launch plan. Have a date? Get a detailed timeline. Still planning? Enter 'TBD' or a quarter such as 'Q2 2027' for a flexible plan. |
| 3 | customer_interview_kit | Customer Interview Kit | Generate interview guides that ADAPT based on interview type, industry, and product complexity. Includes synthesis templates. |
| 4 | retention_playbook | Retention Playbook | Generate retention strategies. Has DISCOVERY MODE: if you don't know WHY people churn, get a churn analysis framework first. |
| 5 | partner_architect | Partner Architect | Design partner programs that ADAPT based on partner model type. Different structures for resellers vs referrals vs integrations vs affiliates. |
| 6 | crisis_planner | Crisis Planner | Generate crisis playbooks. Know your risks? Get specific playbooks. Not sure what to plan for? The tool uses a default set of common crises for your industry (not ranked by likelihood). |
| 7 | competitive_intel | Competitive Intel | Generate battle cards. If you know your strengths/weaknesses, get complete battle cards. If you only know competitors and win/loss stories, we'll derive your positioning. |
| 8 | craft_gtm_analyzer | CRAFT GTM Analyzer | Analyze a GTM document against the CRAFT framework. Parses the content, identifies gaps, scores each dimension and suggests sections to add. |
#### 1. PMF Scorecard (pmf_scorecard)
| Input | Required | Type | Description |
|---|---|---|---|
product | Yes | string | Product name and brief description |
target_market | Yes | one of: enterprise_saas, smb_saas, consumer, marketplace, fintech, logistics_tech, vertical_saas, ai_native, ites, telecom, software, cybersecurity, other | Target market segment (e.g., 'Enterprise SaaS', 'SMB', 'Consumer') |
current_metrics | Yes | string | Your current metrics (will be PARSED). Include any of: MRR, ARR, churn rate, NPS, CAC, LTV, retention rate, activation rate, DAU/MAU, trial conversion, revenue growth. Example: 'MRR: $50K, Churn: 3%, NPS: 45, CAC: $500, LTV: $3000, Retention: 92%' |
time_in_market | No | one of: pre_launch, 0_6_months, 6_12_months, 1_2_years, 2_plus_years | How long the product has been in market (shown in the scorecard) |
customer_feedback | No | string | Optional: Qualitative feedback themes (e.g., 'Users love X but struggle with Y') |
your_sector | No | one of: logistics_tech, fintech, saas, vertical_saas, ai_native, ites, telecom, software, cybersecurity, other | Optional: your company's own industry (not your customers'). When given, it decides the sector notes; when left out, the tool reads it from your words and stays neutral if they are unclear. |
#### 2. Launch Commander (launch_commander)
| Input | Required | Type | Description |
|---|---|---|---|
product_feature | Yes | string | What you're launching (product/feature name and description) |
launch_type | Yes | one of: major_release, feature_launch, beta_launch, product_update, market_expansion | Type of launch determines plan complexity |
target_segments | Yes | string | Target customer segments (comma-separated) |
goals | Yes | string | Launch success metrics (e.g., '500 signups, $50k pipeline, 10% trial conversion, 1000 downloads') |
launch_date | No | string | Target launch date. Accepts: 'YYYY-MM-DD', 'Q1 2027', 'March 2027', or 'TBD' for planning mode |
available_channels | No | string | Optional: Marketing channels available (comma-separated). E.g., 'email, linkedin, blog, webinar, PR, paid_ads, community' |
team_size | No | one of: solo, small_2_5, medium_6_15, large_15_plus | Marketing/GTM team size affects task distribution |
budget_level | No | one of: bootstrap, moderate, well_funded | Budget affects recommended tactics |
your_sector | No | one of: logistics_tech, fintech, saas, vertical_saas, ai_native, ites, telecom, software, cybersecurity, other | Optional: your company's own industry (not your customers'). When given, it decides the sector notes; when left out, the tool reads it from your words and stays neutral if they are unclear. |
#### 3. Customer Interview Kit (customer_interview_kit)
| Input | Required | Type | Description |
|---|---|---|---|
interview_type | Yes | one of: discovery, validation, feedback, churn, win_loss, persona_research | Type of interview determines question focus |
product_context | Yes | string | Product/service being researched |
target_persona | Yes | string | Who you're interviewing (role/title) |
industry | No | one of: saas, fintech, logistics_tech, vertical_saas, ai_native, ites, telecom, software, cybersecurity, ecommerce, marketplace, enterprise_software, consumer, other | Industry affects terminology and context |
product_complexity | No | one of: simple, moderate, complex, highly_technical | Affects technical depth of questions |
key_hypotheses | No | string | Optional: Hypotheses to validate during interview |
your_sector | No | one of: logistics_tech, fintech, saas, vertical_saas, ai_native, ites, telecom, software, cybersecurity, other | Optional: your company's own industry (not your customers'). When given, it decides the sector notes; when left out, the tool reads it from your words and stays neutral if they are unclear. |
#### 4. Retention Playbook (retention_playbook)
| Input | Required | Type | Description |
|---|---|---|---|
customer_segment | Yes | string | Customer segment to focus on |
business_model | Yes | one of: saas_subscription, usage_based, marketplace, transactional, freemium, enterprise_contract | Business model affects health score weighting |
current_churn_rate | Yes | string | Current churn rate (e.g., '5%' or '5% monthly') |
churn_reasons | No | string | OPTIONAL: Known churn reasons (comma-separated). If you don't know, leave blank to get discovery mode with churn analysis framework |
available_data_signals | No | string | What usage data you can track (comma-separated). E.g., 'login frequency, feature usage, support tickets, NPS responses' |
cs_team_size | No | one of: no_dedicated_cs, small_1_3, medium_4_10, large_10_plus | CS team capacity affects intervention strategy |
current_interventions | No | string | Optional: What retention tactics you already do |
your_sector | No | one of: logistics_tech, fintech, saas, vertical_saas, ai_native, ites, telecom, software, cybersecurity, other | Optional: your company's own industry (not your customers'). When given, it decides the sector notes; when left out, the tool reads it from your words and stays neutral if they are unclear. |
#### 5. Partner Architect (partner_architect)
| Input | Required | Type | Description |
|---|---|---|---|
company | Yes | string | Your company name |
product | Yes | string | Product partners will sell/integrate |
partner_model | Yes | one of: reseller, referral, integration_tech, agency_si, affiliate, oem_white_label | Partner type determines program structure |
partner_goals | Yes | string | Revenue/growth targets from partners |
your_deal_size | Yes | string | Average deal size as one amount (e.g., '$5000 ACV', '$5K' or '$500/month'; a range is refused). It scales the example commission amounts; the example rates are fixed |
partner_support_capacity | No | one of: minimal_self_serve, moderate, high_touch | How much partner support can you provide? |
existing_partners | No | string | Optional: Current partner types/count |
your_sector | No | one of: logistics_tech, fintech, saas, vertical_saas, ai_native, ites, telecom, software, cybersecurity, other | Optional: your company's own industry (not your customers'). When given, it decides the sector notes; when left out, the tool reads it from your words and stays neutral if they are unclear. |
#### 6. Crisis Planner (crisis_planner)
| Input | Required | Type | Description |
|---|---|---|---|
company | Yes | string | Company name |
industry | Yes | one of: fintech, saas, logistics_tech, vertical_saas, ai_native, ites, telecom, software, cybersecurity, ecommerce, enterprise, consumer, other | Industry affects which crises to prioritize |
customer_base | Yes | one of: b2b_enterprise, b2b_smb, b2c_consumer, mixed | Customer type affects communication approach |
data_sensitivity | Yes | one of: high_pii_financial, medium_business_data, low_general | Data sensitivity affects security protocols |
potential_crises | No | string | OPTIONAL: Crisis types to plan for (comma-separated). If not provided, the tool uses a default set of common crises for your industry (not ranked by likelihood). Options: data_breach, service_outage, pr_incident, executive_departure, security_vulnerability, regulatory_action, product_safety, customer_data_exposure |
company_size | No | one of: startup_under_50, scaleup_50_200, midsize_200_1000, enterprise_1000_plus | Affects response team structure |
compliance_requirements | No | string | Optional: Relevant compliance, comma-separated (for example ISO 27001, SOC 2, GDPR). Each item is named in the notification steps |
your_sector | No | one of: logistics_tech, fintech, saas, vertical_saas, ai_native, ites, telecom, software, cybersecurity, other | Optional: your company's own industry (not your customers'). When given, it decides the sector notes; when left out, the tool reads it from your words and stays neutral if they are unclear. |
#### 7. Competitive Intel (competitive_intel)
| Input | Required | Type | Description |
|---|---|---|---|
your_product | Yes | string | Your product name and brief description |
competitors | Yes | string | Competitor names (comma-separated). Will generate battle card for EACH |
your_strengths | No | string | OPTIONAL: What you do better (comma-separated). Will be DERIVED from wins/losses if not provided |
your_weaknesses | No | string | OPTIONAL: Where competitors beat you (comma-separated). Will be DERIVED from wins/losses if not provided |
competitor_details | No | string | Optional: Any known details about competitors. E.g., 'Competitor A is cheaper, Competitor B targets enterprise' |
common_objections | No | string | Sales objections you hear (comma-separated). E.g., 'too expensive, missing X feature, competitor has better Y' |
recent_wins | No | string | Why customers chose you over competitors: will be used to DERIVE strengths |
recent_losses | No | string | Why you lost deals to competitors: will be used to DERIVE weaknesses |
your_sector | No | one of: logistics_tech, fintech, saas, vertical_saas, ai_native, ites, telecom, software, cybersecurity, other | Optional: your company's own industry (not your customers'). When given, it decides the sector notes; when left out, the tool reads it from your words and stays neutral if they are unclear. |
#### 8. CRAFT GTM Analyzer (craft_gtm_analyzer)
| Input | Required | Type | Description |
|---|---|---|---|
document_content | Yes | string | The GTM document/plan to analyze. Paste full content: it will be PARSED and EVALUATED |
document_type | Yes | one of: gtm_strategy, launch_plan, campaign_brief, quarterly_plan, project_proposal, marketing_plan | Type of document (shown in the analysis) |
intended_audience | No | string | Optional: Who will read/approve this document |
desired_outcome | No | string | Optional: What action should this document drive |
your_sector | No | one of: logistics_tech, fintech, saas, vertical_saas, ai_native, ites, telecom, software, cybersecurity, other | Optional: your company's own industry (not your customers'). When given, it decides the sector notes; when left out, the tool reads it from your words and stays neutral if they are unclear. |
Shashwat Ghosh, Co-Founder and Fractional CMO, Helix GTM Consulting, with 24+ years in B2B and 10+ years of fractional experience
MIT
The same tools are also available as a hosted MCP server, so they work in Claude on the web, desktop and mobile without installing anything.
https://craft-gtm.gtmhelix.com/mcpclaude mcp add --transport http craft-gtm https://craft-gtm.gtmhelix.com/mcpThe npm package (stdio) and the hosted server run the same createServer() code in src/server.ts.
Full policy: <https://craft-gtm.gtmhelix.com/privacy/> (also in PRIVACY.md).
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.