Skip to content

AgentAegis MCP server

io.github.astafford8488/agentaegisastafford8488streamable-httpv0.3.0Scanned

Pay-per-call cybersecurity for AI agents: vuln scans, threat intel, compliance, code security.

28 tools: 21 change things and 7 only read.

Tools: 0 delete, 0 irreversible, 21 write, 7 read.
  • Reads private data · 1
  • Reads text others write · 14
  • Can send data out · 7
  • 6 findings

Sign-in required, no OAuth metadata. Remote posture

Anlyon alternative

21 of these tools change something. Run those as governed actions.

Declare what an action changes, and Anlyon puts it under a shared limit, binds every approval to the exact request, and grades the outcome. It works on any HTTPS API you can describe as an action.

Anlyon Radar does not run, proxy or govern this MCP server. A tool your own code calls directly is not routed through Anlyon and is not governed by it.

Import an OpenAPI spec →

What the AgentAegis MCP server can do

The AgentAegis MCP server is published by the GitHub account astafford8488. It has 28 tools. It can create or change data (21 tools: compliance_framework_check, evidence_collect, control_gap_analysis, audit_report_generate and 17 more), but cannot delete data. 7 tools only read.

1 tool reads private data (threat_intel_lookup); 14 tools bring in text other people can write (vuln_scan_network, vuln_scan_web_app, cve_lookup and 11 more); 7 tools can send data out of your system (vuln_scan_web_app, sast_scan, secret_scan and 4 more). On its own it has all three parts of a prompt-injection exfiltration path: private data, text others write, and a way out.

The fixed checks matched 6 findings (annotation-conflict and credentials-mention). Each one comes with the text it matched.

Last scanned Oct 21, 2026, version 0.3.0.

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.

28 tools

Changes things 21Claude Code: ask
  • access_reviewReview user and role assignments you supply against least-privilege, flagging excessive, stale or orphaned access.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.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Review user and role assignments you supply against least-privilege, flagging excessive, stale or orphaned access. Analyzes data the caller provides; it does not connect to an identity provider. Costs $1 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • audit_report_generateSynthesize findings into an audit-ready compliance report.1 finding

    Why it is in this group

    • The name "audit_report_generate" says "generate": a word for a change.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    Findings

    • Warn

      The server marks "audit_report_generate" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

      Matched text

      annotations: {"readOnlyHint":true,"idempotentHint":true,"openWorldHint":false}; The name "audit_report_generate" says "generate": a word for a change.

      What this check looks for
    Description, as the server wrote it
    Synthesize findings into an audit-ready compliance report. Use at the END of an engagement, once gaps are closed. If the user only wants to know where they currently stand, run compliance_framework_check instead — it costs less and answers that question directly. Costs $5 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • compliance_framework_checkAssess an organization's security posture against a compliance framework (SOC 2, ISO 27001, HIPAA, PCI-DSS, NIST CSF) and report per-control status.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.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Assess an organization's security posture against a compliance framework (SOC 2, ISO 27001, HIPAA, PCI-DSS, NIST CSF) and report per-control status. Use this FIRST when asked whether the org is audit-ready; control_gap_analysis builds on its output. Costs $2 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • control_gap_analysisTurn unmet compliance controls into a prioritized remediation roadmap with effort estimates.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.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Turn unmet compliance controls into a prioritized remediation roadmap with effort estimates. Use after compliance_framework_check to answer 'what do we fix first'. Costs $2 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • credential_checkCheck whether an email address or domain appears in known credential-breach corpora (Have I Been Pwned), with the breaches and data classes exposed.Reads text others write1 finding

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.

    Findings

    • Info

      The description mentions "credential".

      Matched text

      description: "...mail address or domain appears in known credential-breach corpora (Have I Been Pwned), wit..."

      What this check looks for
    Description, as the server wrote it
    Check whether an email address or domain appears in known credential-breach corpora (Have I Been Pwned), with the breaches and data classes exposed. Use when assessing account-takeover exposure. Costs $2 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: true
  • dependency_auditAudit a dependency manifest or https git repo for known-vulnerable packages (trivy): npm, pip, Go, Ruby, Java, Cargo.Reads text others writeCan send data outNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • outboundThe input takes a URL (input.source.url): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    Audit a dependency manifest or https git repo for known-vulnerable packages (trivy): npm, pip, Go, Ruby, Java, Cargo. The cheapest, highest-signal first step when assessing an unfamiliar repository. Costs $2 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: true
  • dns_security_checkCheck a domain's DNS security records — SPF, DKIM, DMARC, DNSSEC — and grade the configuration.Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Check a domain's DNS security records — SPF, DKIM, DMARC, DNSSEC — and grade the configuration. Passive. Covers the records themselves; for full spoofability posture use email_security_audit. Costs $1 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: true
  • email_security_auditFull email-security posture for a domain: whether mail from it can be spoofed, with DMARC/SPF/DKIM alignment and policy strength.Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Full email-security posture for a domain: whether mail from it can be spoofed, with DMARC/SPF/DKIM alignment and policy strength. A superset of dns_security_check for the email question specifically. Costs $2 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: true
  • evidence_collectBuild an evidence-collection plan for specific compliance controls: what artifact each control needs, where it comes from, and what makes it sufficient.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.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Build an evidence-collection plan for specific compliance controls: what artifact each control needs, where it comes from, and what makes it sufficient. Use when preparing for a real audit, after the gaps are known. Plans the collection; does not gather evidence for you. Costs $1 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • incident_triageClassify a security incident and produce severity, likely category, containment steps and a response plan.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.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Classify a security incident and produce severity, likely category, containment steps and a response plan. Use when something has already happened. If all you have is a suspicious IP or domain, run threat_intel_lookup first — it is cheaper and may settle the question. Costs $3 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • mfa_auditAssess MFA coverage and factor strength across a user or configuration set you supply, flagging unenrolled accounts and weak factors such as SMS.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.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Assess MFA coverage and factor strength across a user or configuration set you supply, flagging unenrolled accounts and weak factors such as SMS. Analyzes data the caller provides; it does not connect to an identity provider. Costs $1 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • policy_generateGenerate a tailored written security policy (incident response, access control, encryption, vendor management, remote work, and similar).1 finding

    Why it is in this group

    • The name "policy_generate" says "generate": a word for a change.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    Findings

    • Warn

      The server marks "policy_generate" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

      Matched text

      annotations: {"readOnlyHint":true,"idempotentHint":true,"openWorldHint":false}; The name "policy_generate" says "generate": a word for a change.

      What this check looks for
    Description, as the server wrote it
    Generate a tailored written security policy (incident response, access control, encryption, vendor management, remote work, and similar). Use when a control gap specifically calls for documented policy. Costs $2 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • sast_scanStatic analysis of source code or an https git repo for security flaws (Semgrep): injection, unsafe deserialization, path traversal, crypto misuse.Reads text others writeCan send data out1 finding

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • outboundThe input takes a URL (input.source.url): a request to a URL can carry data to whoever runs it.

    Findings

    • Info

      The description mentions "credentials".

      Matched text

      description: "...C flaws — use secret_scan for hardcoded credentials and dependency_audit for vulnerable pac..."

      What this check looks for
    Description, as the server wrote it
    Static analysis of source code or an https git repo for security flaws (Semgrep): injection, unsafe deserialization, path traversal, crypto misuse. Python, JS/TS, Java, Go, Ruby, PHP, C/C++. For code LOGIC flaws — use secret_scan for hardcoded credentials and dependency_audit for vulnerable packages. Costs $5 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: true
  • scan_mcp_pluginScan an MCP server (git repo or code) for supply-chain risk BEFORE trusting it — exfiltration (secrets/env to the network), prompt-injection sinks, dangerous capabilities, npm inst…Reads text others writeCan send data outNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • outboundThe input takes a URL (input.source.url): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    Scan an MCP server (git repo or code) for supply-chain risk BEFORE trusting it — exfiltration (secrets/env to the network), prompt-injection sinks, dangerous capabilities, npm install hooks, obfuscation, plus Semgrep + secret scanning → a PROCEED/CAUTION/BLOCK verdict with findings. Costs $5 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: true
  • scan_skillScan an agent SKILL (git repo or SKILL.md) for supply-chain risk BEFORE trusting it — prompt-injection / hidden-unicode in the instructions (hard block), over-broad allowed-tools g…Reads text others writeCan send data outNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • outboundThe input takes a URL (input.source.url): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    Scan an agent SKILL (git repo or SKILL.md) for supply-chain risk BEFORE trusting it — prompt-injection / hidden-unicode in the instructions (hard block), over-broad allowed-tools grants, plus exfiltration, dangerous capabilities, secrets and obfuscation in bundled scripts → a PROCEED/CAUTION/BLOCK verdict. Costs $5 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: true
  • secret_scanDetect hardcoded credentials, API keys and tokens in source code or an https git repo (trufflehog), verified against the issuing provider where supported.Reads text others writeCan send data out1 finding

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • outboundThe input takes a URL (input.source.url): a request to a URL can carry data to whoever runs it.

    Findings

    • Info

      The description mentions "credentials".

      Matched text

      description: "Detect hardcoded credentials, API keys and tokens in source code or..."

      What this check looks for
    Description, as the server wrote it
    Detect hardcoded credentials, API keys and tokens in source code or an https git repo (trufflehog), verified against the issuing provider where supported. Use when the question is 'did we commit a secret'. Costs $2 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: true
  • ssl_tls_auditAudit a domain's TLS configuration (sslyze): certificate validity and expiry, protocol versions, cipher suites, and known TLS weaknesses.Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Audit a domain's TLS configuration (sslyze): certificate validity and expiry, protocol versions, cipher suites, and known TLS weaknesses. Passive — safe against any host. Costs $1 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: true
  • vet_endpointComposite trust verdict (PROCEED/CAUTION/BLOCK) for an endpoint an agent is about to call or pay — combines TLS/cert health, DNS hygiene, threat-intel reputation, and domain age in…Reads text others writeCan send data outNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • outboundThe input takes a URL (input.endpoint): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    Composite trust verdict (PROCEED/CAUTION/BLOCK) for an endpoint an agent is about to call or pay — combines TLS/cert health, DNS hygiene, threat-intel reputation, and domain age into one decision with reasons. Prefer this over running ssl_tls_audit + dns_security_check + threat_intel_lookup separately: it costs less than the sum and returns a decision rather than three reports to reconcile. Costs $3 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: true
  • vuln_prioritizeRank vulnerabilities you already have by exploitability and business impact, and group them into remediation actions.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.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Rank vulnerabilities you already have by exploitability and business impact, and group them into remediation actions. Analyzes findings you supply; it discovers nothing on its own. Costs $1 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • vuln_scan_networkDiscover open ports, running services and known vulnerabilities on an IP or domain (nmap).Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Discover open ports, running services and known vulnerabilities on an IP or domain (nmap). SENDS REAL TRAFFIC to the target and may trigger intrusion detection — only run against hosts the caller owns or is explicitly authorized to test, and confirm that first. Pass async:true to get a job_id to poll instead of blocking. Costs $3 per call.

    Annotations the server declares

    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
  • vuln_scan_web_appScan a web application for OWASP Top 10 issues and known CVEs (Nuclei).Reads text others writeCan send data out1 finding

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • outboundThe input takes a URL (input.target_url): a request to a URL can carry data to whoever runs it.

    Findings

    • Info

      The input schema at properties.authentication.required[1] mentions "credentials".

      Matched text

      inputSchema.properties.authentication.required[1]: "credentials"

      What this check looks for
    Description, as the server wrote it
    Scan a web application for OWASP Top 10 issues and known CVEs (Nuclei). SENDS REAL TRAFFIC to the target — authorized targets only, confirm before calling. Pass async:true to get a job_id to poll instead of blocking. Costs $5 per call.

    Annotations the server declares

    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
Only reads 7Show 7 read-only tools
  • account_balanceReturns the calling API key's prepaid balance, monthly limit, current month usage, and a breakdown of how many of each tool the customer can still afford.No findings

    Why it is in this group

    • The description's first verb is "return": a word for reading.
    Description, as the server wrote it
    Returns the calling API key's prepaid balance, monthly limit, current month usage, and a breakdown of how many of each tool the customer can still afford. Free to call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • agent_historyLists your recent scans (scan_id, tool, target, status, time) so you can retrieve or chain from a prior result.No findings

    Why it is in this group

    • The description's first verb is "list": a word for reading.
    Description, as the server wrote it
    Lists your recent scans (scan_id, tool, target, status, time) so you can retrieve or chain from a prior result. Optional limit/tool/target/since filters. Free to call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • agent_scan_getRetrieves one of your prior scans by scan_id, including the stored full output, so you can build on earlier results without re-paying.No findings

    Why it is in this group

    • The name "agent_scan_get" says "get": a word for reading.
    Description, as the server wrote it
    Retrieves one of your prior scans by scan_id, including the stored full output, so you can build on earlier results without re-paying. Free to call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • agent_whoamiReturns your persistent AgentAegis agent identity (agent_id), how you're identified (API key / wallet / anonymous session), and lifetime call count + spend.No findings

    Why it is in this group

    • The name "agent_whoami" says "whoami": a word for reading.
    Description, as the server wrote it
    Returns your persistent AgentAegis agent identity (agent_id), how you're identified (API key / wallet / anonymous session), and lifetime call count + spend. Free to call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • cve_lookupLook up one CVE by identifier: CVSS score and vector, affected products, patch availability and references.Reads text others writeNo findings

    Why it is in this group

    • The name "cve_lookup" says "lookup": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Look up one CVE by identifier: CVSS score and vector, affected products, patch availability and references. Use when a specific CVE ID is already known. Costs $1 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: true
  • helpReturns AgentAegis FAQ — authentication, balance/billing, tool catalog, async jobs, error codes, x402, rate limits, security.No findings

    Why it is in this group

    • The description's first verb is "return": a word for reading.
    Description, as the server wrote it
    Returns AgentAegis FAQ — authentication, balance/billing, tool catalog, async jobs, error codes, x402, rate limits, security. Optional topic filter. Free to call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: false
  • threat_intel_lookupReputation and indicator lookup for an IP or domain across AbuseIPDB, AlienVault OTX and abuse.ch.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "threat_intel_lookup" says "lookup": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "vault": it reads secrets.
    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Reputation and indicator lookup for an IP or domain across AbuseIPDB, AlienVault OTX and abuse.ch. The cheapest way to check whether an indicator is known-bad. Interpret with care: large CDN, cloud and payment infrastructure routinely returns reputation hits, so only a curated active-malware hit is strong evidence on its own. Costs $2 per call.

    Annotations the server declares

    • readOnlyHint: true
    • idempotentHint: true
    • openWorldHint: true

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.

98 paths in this server alone from private data, through text someone else wrote, to a way out. Denying 1 private-data tool breaks every one.

Reads private data1
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write14
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out7
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
deny to break every path
{
  "permissions": {
    "deny": [
      "mcp__agentaegis__threat_intel_lookup"
    ]
  }
}

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.

Sign-in required, no OAuth metadata

Looked at Nov 20, 2026, 19:02 UTC, unchanged since Nov 6, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
404
tools/list served without credentials
no
Tools that change things, listed without credentials
0
Protocol version
not observed
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

The probe stopped: initialize answered HTTP 404.

Posture changes

  1. Nov 6, 2026
    • HTTP status of an anonymous initialize: 200 to 404
    • tools/list served without credentials: yes to no
    • Tools that change things, listed without credentials: 21 to 0
    • Protocol version: 2025-06-18 to none

Findings

Matches of fixed checks, with the text they matched.

2 warn4 info
  • The server marks "audit_report_generate" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

    Matched text

    annotations: {"readOnlyHint":true,"idempotentHint":true,"openWorldHint":false}; The name "audit_report_generate" says "generate": a word for a change.

    What this check looks for
  • The server marks "policy_generate" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

    Matched text

    annotations: {"readOnlyHint":true,"idempotentHint":true,"openWorldHint":false}; The name "policy_generate" says "generate": a word for a change.

    What this check looks for
  • The input schema at properties.authentication.required[1] mentions "credentials".

    Matched text

    inputSchema.properties.authentication.required[1]: "credentials"

    What this check looks for
  • The description mentions "credentials".

    Matched text

    description: "...C flaws — use secret_scan for hardcoded credentials and dependency_audit for vulnerable pac..."

    What this check looks for
  • The description mentions "credentials".

    Matched text

    description: "Detect hardcoded credentials, API keys and tokens in source code or..."

    What this check looks for
  • The description mentions "credential".

    Matched text

    description: "...mail address or domain appears in known credential-breach corpora (Have I Been Pwned), wit..."

    What this check looks for

History

Changes to its tool definitions and to its remote endpoint, newest first.

  1. PostureRemote posture changed. HTTP status of an anonymous initialize: 200 -> 404; tools/list served without credentials: yes -> no; Tools that change things, listed without credentials: 21 -> 0 (and 1 more).View diff
    • HTTP status of an anonymous initialize: 200 to 404
    • tools/list served without credentials: yes to no
    • Tools that change things, listed without credentials: 21 to 0
    • Protocol version: 2025-06-18 to none

Install

Pinned to version 0.3.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http agentaegis https://agentaegis-mcp-production.up.railway.app/mcp
claude mcp add --transport http agentaegis https://agentaegis-mcp-production.up.railway.app/mcp

Run 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.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__agentaegis__cve_lookup",
      "mcp__agentaegis__threat_intel_lookup",
      "mcp__agentaegis__account_balance",
      "mcp__agentaegis__help",
      "mcp__agentaegis__agent_whoami",
      "mcp__agentaegis__agent_history",
      "mcp__agentaegis__agent_scan_get"
    ],
    "ask": [
      "mcp__agentaegis__compliance_framework_check",
      "mcp__agentaegis__evidence_collect",
      "mcp__agentaegis__control_gap_analysis",
      "mcp__agentaegis__audit_report_generate",
      "mcp__agentaegis__policy_generate",
      "mcp__agentaegis__vuln_scan_network",
      "mcp__agentaegis__vuln_scan_web_app",
      "mcp__agentaegis__vuln_prioritize",
      "mcp__agentaegis__ssl_tls_audit",
      "mcp__agentaegis__sast_scan",
      "mcp__agentaegis__secret_scan",
      "mcp__agentaegis__dependency_audit",
      "mcp__agentaegis__incident_triage",
      "mcp__agentaegis__dns_security_check",
      "mcp__agentaegis__email_security_audit",
      "mcp__agentaegis__access_review",
      "mcp__agentaegis__mfa_audit",
      "mcp__agentaegis__credential_check",
      "mcp__agentaegis__vet_endpoint",
      "mcp__agentaegis__scan_mcp_plugin",
      "mcp__agentaegis__scan_skill"
    ],
    "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 agentaegis. Review it before you use it.

Capability tags behind 14 rules
  • mcp__agentaegis__vuln_scan_network untrusted-input
  • mcp__agentaegis__vuln_scan_web_app untrusted-input, outbound
  • mcp__agentaegis__cve_lookup untrusted-input
  • mcp__agentaegis__ssl_tls_audit untrusted-input
  • mcp__agentaegis__sast_scan untrusted-input, outbound
  • mcp__agentaegis__secret_scan untrusted-input, outbound
  • mcp__agentaegis__dependency_audit untrusted-input, outbound
  • mcp__agentaegis__threat_intel_lookup private-data, untrusted-input
  • mcp__agentaegis__dns_security_check untrusted-input
  • mcp__agentaegis__email_security_audit untrusted-input
  • mcp__agentaegis__credential_check untrusted-input
  • mcp__agentaegis__vet_endpoint untrusted-input, outbound
  • mcp__agentaegis__scan_mcp_plugin untrusted-input, outbound
  • mcp__agentaegis__scan_skill untrusted-input, outbound

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.astafford8488/agentaegis/verify?hash=51c7da112fb23ed245e248b43b82ccd20df1a2978ed4350bb12c9a4aa04ce4c7" | jq -e '.changed == false'

Pinned hash 51c7da112fb2, from Oct 21, 2026, 19:06 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/io.github.astafford8488/agentaegis/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for io.github.astafford8488/agentaegis
[![Anlyon Radar](https://radar.anlyon.com/badge/io.github.astafford8488/agentaegis.svg)](https://radar.anlyon.com/servers/io.github.astafford8488/agentaegis)

Source: MCP Registry · repository. Synced Oct 21, 2026, 18:23 UTC.

AgentAegis MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the AgentAegis MCP server safe?

No scan can promise that. What its definitions show: 28 tools, 21 of which can change, send or delete something, and 6 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 AgentAegis MCP server delete data?

No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.

Can the AgentAegis 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 AgentAegis MCP server have?

28 tools: 21 write, 7 read. compliance_framework_check, evidence_collect, control_gap_analysis, audit_report_generate, policy_generate, vuln_scan_network, vuln_scan_web_app, vuln_prioritize and 20 more.

Does the AgentAegis MCP server need sign-in or OAuth?

Its remote endpoint did not answer the last anonymous look.

Does the AgentAegis MCP server send data to outside services?

7 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): vuln_scan_web_app, sast_scan, secret_scan, dependency_audit, vet_endpoint, scan_mcp_plugin and 1 more.

Has the AgentAegis MCP server changed its tools?

Not since Anlyon Radar started tracking it. Last scanned Oct 21, 2026.