Is "@sanity/agent-context-explorer" on MCP Registry Safe to Install?

torbratsbergsanity · mcp · v0.0.4

Exploration tool for Sanity Agent Context — produces knowledge docs for production AI agents

Risk Assessment

Analyzed
89.04
out of 100
CRITICAL

55 security findings detected across all analyzers

MCP server analyzed for tool poisoning, prompt injection, and data exfiltration

Severity Breakdown

1
Critical
18
High
34
Medium
0
Low
0
Info

Finding Categories

10
Malware Signatures
4
Secrets
32
IoC Indicators

YARA Rules Matched

4 rules(10 hits)
credential env files postinstall file manipulation postinstall network communication postinstall system command

MCP Server Analysis

MCP servers expose tools and resources to AI assistants. Unlike browser extensions, they run as standalone processes with direct system access. Tool definitions are analyzed for prompt injection, data exfiltration, and tool poisoning patterns.

About This Extension

Exploration tool for Sanity Agent Context — produces knowledge docs for production AI agents

Detailed Findings

23 total

YARA Rule Matches

4 rules

Indicators of Compromise

Network indicators, suspicious strings, and potential IoCs extracted during analysis

URLs
4
IP Addresses
1
Domains
26
Strings
32

All Indicators · 32

Domain
detected Domain: priorfindings.map

XIOC detected Domain: priorfindings.map

extracted_from_files

IP
detected IP: ::

XIOC detected IP: ::

extracted_from_files

Domain
detected Domain: error.issues.map

XIOC detected Domain: error.issues.map

extracted_from_files

Domain
detected Domain: cli.js.map

XIOC detected Domain: cli.js.map

extracted_from_files

Domain
detected Domain: tool.name

XIOC detected Domain: tool.name

extracted_from_files

Domain
detected Domain: sanity.io

XIOC detected Domain: sanity.io

extracted_from_files

Other
detected Email: [email protected]

XIOC detected Email: [email protected]

extracted_from_files

URL
detected URL: https://github.com/sanity-io/agent-context).

XIOC detected URL: https://github.com/sanity-io/agent-context).

extracted_from_files

Domain
detected Domain: attempt.toolcalls.map

XIOC detected Domain: attempt.toolcalls.map

extracted_from_files

Domain
detected Domain: mcptool.name

XIOC detected Domain: mcptool.name

extracted_from_files

Domain
detected Domain: result.question.id

XIOC detected Domain: result.question.id

extracted_from_files

Domain
detected Domain: validated.questions.map

XIOC detected Domain: validated.questions.map

extracted_from_files

Domain
detected Domain: q.id

XIOC detected Domain: q.id

extracted_from_files

Domain
detected Domain: err.issues.map

XIOC detected Domain: err.issues.map

extracted_from_files

Domain
detected Domain: values.help

XIOC detected Domain: values.help

extracted_from_files

URL
detected URL: https://api.sanity.io/vX/agent-context/PROJECT_ID/DATASET/SLUG

XIOC detected URL: https://api.sanity.io/vX/agent-context/PROJECT_ID/DATASET/SLUG

extracted_from_files

Domain
detected Domain: calllogs.map

XIOC detected Domain: calllogs.map

extracted_from_files

Domain
detected Domain: this.name

XIOC detected Domain: this.name

extracted_from_files

Domain
detected Domain: date.now

XIOC detected Domain: date.now

extracted_from_files

Domain
detected Domain: sections.map

XIOC detected Domain: sections.map

extracted_from_files

Domain
detected Domain: tool2.name

XIOC detected Domain: tool2.name

extracted_from_files

Domain
detected Domain: stats.total

XIOC detected Domain: stats.total

extracted_from_files

Domain
detected Domain: github.com

XIOC detected Domain: github.com

extracted_from_files

Domain
detected Domain: api.sanity.io

XIOC detected Domain: api.sanity.io

extracted_from_files

Domain
detected Domain: exploration-results.md

XIOC detected Domain: exploration-results.md

extracted_from_files

Domain
detected Domain: batch.map

XIOC detected Domain: batch.map

extracted_from_files

Domain
detected Domain: question.id

XIOC detected Domain: question.id

extracted_from_files

Domain
detected Domain: tools.map

XIOC detected Domain: tools.map

extracted_from_files

Domain
detected Domain: t.name

XIOC detected Domain: t.name

extracted_from_files

URL
detected URL: https://github.com/sanity-io/agent-context))

XIOC detected URL: https://github.com/sanity-io/agent-context))

extracted_from_files

URL
detected URL: https://api.sanity.io/vX/agent-context/YOUR_PROJECT_ID/YOUR_DATASET/YOUR_SLUG

XIOC detected URL: https://api.sanity.io/vX/agent-context/YOUR_PROJECT_ID/YOUR_DATASET/YOUR_SLUG

extracted_from_files

Domain
detected Domain: args.help

XIOC detected Domain: args.help

extracted_from_files

Security Analysis Summary

Security Analysis Overview

@sanity/agent-context-explorer is a mcp extension published by torbratsbergsanity. Version 0.0.4 has been analyzed by the Risky Plugins security platform, receiving a risk score of 89.04/100 (CRITICAL risk) based on 55 security findings.

Risk Assessment

This extension presents critical security risk. Severe issues were detected, potentially including malware indicators, exposed secrets, or dangerous behaviors. Installation is strongly discouraged until these issues are addressed.

Findings Breakdown

  • Critical: 1 finding(s)
  • High: 18 finding(s)
  • Medium: 34 finding(s)

What Was Analyzed

The security assessment covers multiple analysis categories:

  • Malware Detection: YARA rule matching against 2,400+ malware signatures
  • Secret Detection: Scanning for exposed API keys, tokens, and credentials
  • Static Analysis: Code-level security analysis for common vulnerability patterns
  • Network Analysis: Detection of suspicious network communications and endpoints
  • Obfuscation Detection: Identification of code obfuscation techniques

Developer Information

@sanity/agent-context-explorer is published by torbratsbergsanity on the mcp marketplace.

Recommendation

This extension is not recommended for installation without thorough manual review. Consider alternatives with lower risk scores, or contact the developer to address the identified security concerns.

Frequently Asked Questions