MCP Registry

@iflow-mcp/mcp-sysoperator

3c0f9091-cfcf-5447-a830-4b51102adbfb | Store v0.1.0
100/ 100
CRITICAL risk
AI review verdict
Needs follow up

From the RiskyPlugins AI security review of the observed evidence.

Analysis record

Analysed
2 months ago 26 August 2026
Version
v0.1.0
Artifact
SHA256 ba156a1a…99812a
Detail
Full list

Evidence ledger

Ranked by severity · findings with a source location link to the code viewer

372 findings

YARA Rule Matches

15 rules
SeverityRuleHitsFilesMetadata
LOWWeakSSLTLSProtocolsShouldNotBeUsed 2
demos/aws-terraform-lamp/ansible/roles/web/templates/httpd.conf.j2demos/aws-terraform-lamp/ansible/roles/web/templates/vhost.conf.j2
-
LOWcredential env files 14
demos/aws-terraform-lamp/terraform/modules/loadbalancing/main.tfdemos/aws-terraform-lamp/terraform/modules/database/main.tfdemos/aws-terraform-lamp/ansible/inventory/aws_ec2.yml +11 more
-
LOWpostinstall persistence mechanism 13
demos/aws-terraform-lamp/ansible/roles/db_client/templates/logrotate-db.j2demos/aws-terraform-lamp/ansible/roles/db_client/tasks/main.ymldemos/aws-terraform-lamp/ansible/roles/app/tasks/main.yml +10 more
-
LOWpostinstall file download 29
demos/aws-terraform-lamp/ansible/roles/app/templates/script.js.j2demos/aws-terraform-lamp/terraform/modules/loadbalancing/main.tfdemos/aws-terraform-lamp/ansible/roles/app/templates/index.php.j2 +26 more
-
LOWcredential aws profile 4
demos/aws-lamp-stack/group_vars/localstack.ymldemos/aws-lamp-stack/group_vars/aws.ymldemos/aws-terraform-lamp/ansible/group_vars/all.yml +1 more
-
LOWcredential ssh keys 1
examples/inventory.ini
-
LOWUsingCommandLineArguments 1
localstack/localstack_aws_operations.mjs
-
LOWpostinstall network communication 66
demos/aws-terraform-lamp/ansible/group_vars/web_servers.ymldemos/aws-terraform-lamp/ansible/roles/db_client/README.mdsrc/sysoperator/index.ts +63 more
-
LOWpostinstall system command 106
demos/aws-terraform-lamp/ansible/roles/db_client/templates/db-connect.php.j2src/sysoperator/operations/aws.tsdemos/aws-terraform-lamp/terraform/modules/monitoring/main.tf +103 more
-
LOWUsingShellInterpreterWhenExecutingOSCommands 13
localstack/localstack_aws_operations.mjslocalstack/mcp_localstack_patch.jslocalstack/run_sample_playbook.mjs +10 more
-
LOWcredential aws credentials 1
demos/aws-terraform-lamp/ansible/roles/common/tasks/main.yml
-
LOWpostinstall environment access 24
src/sysoperator/operations/inventory.tsbuild/sysoperator/operations/terraform.jssrc/sysoperator/operations/playbooks.ts +21 more
-
LOWpostinstall crypto operations 26
demos/aws-terraform-lamp/README.mdbuild/sysoperator/index.jsdemos/aws-terraform-lamp/terraform/modules/storage/main.tf +23 more
-
LOWpostinstall obfuscation 11
demos/aws-terraform-lamp/terraform/modules/dns/main.tfdocker-build-run.shdemos/aws-terraform-lamp/ansible/roles/web/templates/php.ini.j2 +8 more
-
LOWpostinstall file manipulation 48
demos/aws-lamp-stack/test_with_localstack.shbuild/sysoperator/common/types.d.tslocalstack/localstack_test.mjs +45 more
-

Finding Categories

13
Secrets
54
IoC Indicators

YARA Rules Matched

15 rules(359 hits)
WeakSSLTLSProtocolsShouldNotBeUsed credential env files postinstall persistence mechanism postinstall file download credential aws profile credential ssh keys UsingCommandLineArguments postinstall network communication postinstall system command UsingShellInterpreterWhenExecutingOSCommands credential aws credentials postinstall environment access postinstall crypto operations postinstall obfuscation postinstall file manipulation

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.

AI Security Report

AI Security Review

Evidence context: threat category none; evidence quality moderate.

Tool Poisoning Assessment: Zero tool-poisoning findings were detected. The findings_summary shows "tool-poisoning":"0", meaning no hidden AI manipulation directives were identified in tool descriptions. This is significant because tool poisoning is the defining threat for MCP servers.

Credential and Secret Access: The scan detected 13 critical-severity findings in the secret category. However, the findings bundle does not show which specific files or credential types triggered these detections. For an MCP server described as handling "Infrastructure as Code operations (Ansible, Terraform)", some credential access is expected for API key configuration and cloud provider authentication. The critical question is whether these secrets are being harvested for exfiltration or legitimately configured for infrastructure operations.

Network Activity and Exfiltration: The findings_summary shows "network":"0" - no network findings were detected. This is notable because the exfiltration attack pattern requires both credential access AND network calls to unknown domains. The IOC findings include legitimate infrastructure URLs: XIOC-URL-https://apt.releases.hashicorp.com (Terraform package repository), XIOC-URL-http://169.254.169.254/latest/meta-data/instance-id (AWS metadata endpoint), XIOC-URL-https://localstack.cloud/ (AWS mock service), and XIOC-URL-https://github.com/aws/efs-utils (AWS utilities). These align with the package's stated purpose of infrastructure automation.

False Positive Drivers: Several IOC findings are clearly false positives from the XIOC extractor: XIOC-DOMAIN-date.now (JavaScript code misidentified as domain), XIOC-URL-https://$ (malformed URL), XIOC-URL-https://s3. (incomplete URL). The 154 total IOC findings appear to be string extractions from documentation and configuration rather than actual network destinations.

Strongest Counterargument: The 13 critical secret findings combined with zero user count and unknown publisher (chatflowdev) could indicate credential harvesting. However, without network findings showing exfiltration to unknown domains, and without evidence of credential reads targeting sensitive paths (.ssh/, .aws/credentials, .kube/config), this pattern is more consistent with legitimate infrastructure tooling than malicious exfiltration architecture.

Recommendation: Manual code review of the 13 critical secret findings is required to determine if credentials are being harvested beyond the package's stated infrastructure automation purpose.

Key Reasons

  • 13 critical secret findings detected but source files not specified in evidence
  • Zero tool-poisoning findings - no hidden AI manipulation directives
  • Zero network findings - no evidence of credential exfiltration architecture
  • IOC findings are legitimate infrastructure URLs (HashiCorp, AWS, GitHub) or XIOC false positives
  • Package has zero users and unknown publisher - requires additional vetting

False Positive Considerations

  • XIOC-DOMAIN-date.now - JavaScript code misidentified as domain
  • XIOC-URL-https://$ - malformed URL extraction
  • XIOC-URL-https://s3. - incomplete URL extraction
  • Legitimate infrastructure URLs (HashiCorp, AWS metadata, GitHub) flagged as IoCs

Reviewed 2026-04-27 · version 0.1.0; recommended action: runtime analysis; model confidence 65%.

Source Code Viewer

Source Code Not Available

Source code is not available for this version of the extension.

About This Extension

Model Context Protocol server for Infrastructure as Code operations (Ansible, Terraform)

Frequently Asked Questions