COS Monitor MCP server
io.github.internexio/cos-monitorinternexiostreamable-httpv1.0.0Scanned
Scores website pages for communication quality, tracks regressions, surfaces a fix queue.
13 tools: 6 change things and 7 only read.
- Reads private data · 0
- Reads text others write · 0
- Can send data out · 0
- No findings
Tools listed without sign-in. Remote posture
With Anlyon
6 tools here change things. Govern them.
- Anlyon lists the tools and adopts none.
- Each tool you adopt gets approvals and receipts.
- Your agent never holds the key.
A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.
What the COS Monitor MCP server can do
The COS Monitor MCP server is published by the GitHub account internexio. It has 13 tools. It can create or change data (6 tools: execute_template, chat, analyze_full_comms, profile_agent and 2 more), but cannot delete data. 7 tools only read.
None of its tools reads private data, brings in text other people write, or sends data out.
None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.
Its remote endpoint lists its tools without sign-in, and its tools changed once since Anlyon Radar first saw them.
Last scanned Nov 19, 2026, version 1.0.0.
Tools
Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.
13 tools
Changes things 6Claude Code: ask
analyze_full_commsRun all 7 COS frameworks in parallel for comprehensive analysis.No findings
Why it is in this group
- The name "analyze_full_comms" says "analyze": a word for reading.
- The description's first verb is "run": a word for a change.
Description, as the server wrote it Run all 7 COS frameworks in parallel for comprehensive analysis. This is the most thorough analysis option, running: - Core 4: HAPE, Big Five, Strategic Clarity, Sovereign Mind - Extended 3: Persuasion (domain-specific), Platform, Quality Use this when you need complete analysis across all dimensions. Takes longer but provides the most comprehensive view.
audience_profileInfer OCEAN personality profile from an audience description.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
Description, as the server wrote it Infer OCEAN personality profile from an audience description. Maps a free-text target audience description into a structured psychological profile suitable for personalized outreach (cold email, ads, sales messaging). Returns: - OCEAN scores (openness, conscientiousness, extraversion, agreeableness, neuroticism) - ocean_confidence (0.0-1.0) — low when signals are weak - elm_route ("central" | "peripheral" | "mixed") — how the audience processes persuasion - dominant_traits + trait_rationale - dominant_moral_foundations (Moral Foundations Theory) - vulnerability_flags — audiences requiring careful ethics review - recommended_persuasion_principle (Cialdini) + persuasion_rationale Common use: feed a CRM Person/Account description (title, industry, recent signals) to get a psychology-grounded targeting profile for that prospect.
chatHave a conversation with the COS analysis agent.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
Description, as the server wrote it Have a conversation with the COS analysis agent. The agent can help you: - Analyze content interactively - Get recommendations for improvement - Understand framework scores - Configure analysis settings
execute_templateExecute a specific template with provided variables.No findings
Why it is in this group
- The name "execute_template" says "execute": a word for a change.
Description, as the server wrote it Execute a specific template with provided variables. Templates guide the analysis with pre-defined prompts and variable placeholders. First use get_templates to find available templates and their required variables.
optimize_email_for_prospectGenerate or refine a personalized cold email for a CRM prospect.No findings
Why it is in this group
- The description's first verb is "generate": a word for a change.
Description, as the server wrote it Generate or refine a personalized cold email for a CRM prospect. Composite tool: combines audience profiling (OCEAN + Cialdini), optional agent profiling from writing samples, draft generation (if no draft is supplied), and persuasion + platform scoring in a single call. Designed for CRM integrations like Clarify, HubSpot, Salesforce — pass a Person/Account context, get back a draft + scoring. Returns: - audience_profile: OCEAN scores, ELM route, Cialdini principle - agent_profile: prospect's writing style (if samples provided) - draft: generated or echoed email body - draft_was_generated: bool — whether COS generated the draft - persuasion + platform: full scoring breakdowns - rewrites: prioritized rewrite suggestions - one_thing: the single most important next step - cialdini_principle: recommended influence principle
profile_agentProfile an agent's personality from their writing samples.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
Description, as the server wrote it Profile an agent's personality from their writing samples. Analyzes 1-10 writing samples (3-5 recommended) to infer the author's Big Five (OCEAN) personality traits, communication style, strengths, blind spots, and persuasion profile. This is the inverse of content analysis — instead of "is this content effective?", it answers "who is this writer based on how they communicate?"
Only reads 7Show 7 read-only tools
analyze_contentAnalyze content using all 4 COS frameworks in parallel.No findings
Why it is in this group
- The name "analyze_content" says "analyze": a word for reading.
Description, as the server wrote it Analyze content using all 4 COS frameworks in parallel. Returns comprehensive analysis with: - Overall scores (0-10) for each framework - Dimension breakdowns with weights - Specific recommendations for improvement - Cross-framework insights
analyze_frameworkAnalyze content using a single specific framework.No findings
Why it is in this group
- The name "analyze_framework" says "analyze": a word for reading.
Description, as the server wrote it Analyze content using a single specific framework. Faster than full analysis when you only need one perspective. Frameworks: - hape: Engagement Analysis (novelty, relevance, emotional valence) - big_five: Personality Analysis using OCEAN model (openness, conscientiousness, extraversion, agreeableness, neuroticism) - strategic_clarity: Business message clarity (value prop, differentiation, CTA) - framing_strategy: Cognitive frames and power positioning
analyze_persuasionAnalyze content using domain-specific persuasion frameworks.No findings
Why it is in this group
- The name "analyze_persuasion" says "analyze": a word for reading.
Description, as the server wrote it Analyze content using domain-specific persuasion frameworks. Each domain has specialized scoring dimensions: - business: B2B/B2C messaging, ROI framing, objection handling - politics: Political messaging, polarization awareness, coalition building - health: Medical accuracy, safety messaging, behavior change (CRITICAL domain) - masculinity: Identity messaging (neutral Identity module: identity security, meaning preservation, standing safety) - comedy: Humor mechanics, timing, callback patterns
analyze_platformAnalyze content for platform-specific optimization.No findings
Why it is in this group
- The name "analyze_platform" says "analyze": a word for reading.
Description, as the server wrote it Analyze content for platform-specific optimization. Evaluates content against platform constraints and algorithm preferences: - Character limits and formatting rules - Algorithm optimization signals - Engagement pattern recommendations - Platform-specific best practices Supported platforms: twitter, linkedin, email, youtube, tiktok, instagram, facebook, medium, substack, podcast, newsletter, slack, discord
analyze_qualityAnalyze content quality across 5 dimensions.No findings
Why it is in this group
- The name "analyze_quality" says "analyze": a word for reading.
Description, as the server wrote it Analyze content quality across 5 dimensions. Quality dimensions evaluated: - Clarity: Is the message easy to understand? - Coherence: Does the content flow logically? - Correctness: Grammar, spelling, factual accuracy - Completeness: Are all necessary elements present? - Conciseness: Is the content appropriately tight?
get_template_detailsGet detailed information about a specific template.No findings
Why it is in this group
- The name "get_template_details" says "get": a word for reading.
Description, as the server wrote it Get detailed information about a specific template. Returns the template's: - Name and description - Required and optional variables with types - Categories and tags - Scoring dimensions and weights
get_templatesList available analysis templates.No findings
Why it is in this group
- The name "get_templates" says "get": a word for reading.
Description, as the server wrote it List available analysis templates. Templates are pre-configured analysis scenarios for common use cases: - Email outreach optimization - LinkedIn post analysis - Sales pitch review - Content marketing assessment
This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.
Exfil surface
Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.
No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data or untrusted-input or outbound.
- Reads private data0
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write0
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out0
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.
Remote posture
What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.
Looked at Nov 20, 2026, 19:06 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.
- Reachable
- yes
- HTTP status of an anonymous initialize
- 200
- tools/list served without credentials
- yes
- Tools that change things, listed without credentials
- 6
- Protocol version
- 2025-06-18
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- not observed
- Authorization servers
- not observed
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- not observed
- Client ID metadata documents
- not observed
- WarnThe whole server
The endpoint served tools/list without credentials, and the list has 6 tools that change things (6 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://mcp.semalytics.io/mcp: tools/list returned 13 tools to an anonymous client, including execute_template (write), chat (write), analyze_full_comms (write), profile_agent (write), audience_profile (write) and 1 more
Findings
Matches of fixed checks, with the text they matched.
No check matched anything in these tool definitions. That is not a safety guarantee.
History
Changes to its tool definitions and to its remote endpoint, newest first.
Tools1 changed.~1 changedView diff
- Changed:analyze_persuasionOnly readsdescription
Install
Pinned to version 1.0.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http cos-monitor https://mcp.semalytics.io/mcpclaude mcp add --transport http cos-monitor https://mcp.semalytics.io/mcpRun in a terminal.
Claude Code permissions
Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.
{
"permissions": {
"allow": [
"mcp__cos-monitor__analyze_content",
"mcp__cos-monitor__analyze_framework",
"mcp__cos-monitor__get_templates",
"mcp__cos-monitor__get_template_details",
"mcp__cos-monitor__analyze_persuasion",
"mcp__cos-monitor__analyze_platform",
"mcp__cos-monitor__analyze_quality"
],
"ask": [
"mcp__cos-monitor__execute_template",
"mcp__cos-monitor__chat",
"mcp__cos-monitor__analyze_full_comms",
"mcp__cos-monitor__profile_agent",
"mcp__cos-monitor__audience_profile",
"mcp__cos-monitor__optimize_email_for_prospect"
],
"deny": []
}
}Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as cos-monitor. Review it before you use it.
Pin
Keep the decision: fail CI when the tools change, follow every change, or show a badge.
Pin in CI
Fail a build when the tools you reviewed are not the tools the server serves today.
curl -s "https://radar.anlyon.com/api/v1/servers/io.github.internexio/cos-monitor/verify?hash=3b2d02543043b37c32898b2367340a7820202eb84e3e4258d2ab7a69d338911c" | jq -e '.changed == false'Pinned hash 3b2d02543043, from Nov 19, 2026, 21:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/io.github.internexio/cos-monitor/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/io.github.internexio/cos-monitor)Source: MCP Registry · repository. Synced Oct 21, 2026, 18:36 UTC.
COS Monitor MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the COS Monitor MCP server safe?
- No scan can promise that. What its definitions show: 13 tools, 6 of which can change, send or delete something, and 0 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.
Can the COS Monitor MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the COS Monitor MCP server do things that cannot be undone?
- No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.
What tools does the COS Monitor MCP server have?
- 13 tools: 6 write, 7 read. execute_template, chat, analyze_full_comms, profile_agent, audience_profile, optimize_email_for_prospect, analyze_content, analyze_framework and 5 more.
Does the COS Monitor MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the COS Monitor MCP server send data to outside services?
- None of its tools sends, posts or publishes, or takes a recipient, channel or URL.
Has the COS Monitor MCP server changed its tools?
- Yes, once since Anlyon Radar started tracking it, most recently on Nov 19, 2026. Every change is on this page and in its Atom feed.