ip-investigation

Use when a user asks to investigate, check, enrich, or characterize an IP address (IPv4 or IPv6). Chains VirusTotal, Shodan, AbuseIPDB, GreyNoise, OTX, and optionally Censys in parallel, then consolid

SKILL.md

GitHub

ip-investigation

Investigate an IP address end-to-end. Chain multiple lookup-* skills, consolidate results, prioritize follow-up IOCs, and return a rated investigation summary. You invoke skills — you don't call APIs directly.

This skill invokes: /lookup-liberty91 (first-party — run first), /lookup-virustotal, /lookup-shodan, /lookup-abuseipdb, /lookup-greynoise, /lookup-otx, optionally /lookup-censys, optionally /lookup-reversinglabs (network threat-intel + sample fan-out), optionally /lookup-crowdstrike (IOC reputation + actor/malware links), optionally /lookup-misp and/or /lookup-opencti (internal correlation), optionally /lookup-sentinel (exposure check in your own telemetry), then /source-assessment, /tlp-guide, /confidence-levels.

When to invoke

  • User asks to investigate, check, or enrich an IP
  • User pastes an IP and asks "what is this?" / "is this malicious?"
  • Another skill hands off an IP indicator for follow-up
  • IOC triage pipeline

Inputs

indicator: <IPv4 or IPv6 address>
context: <optional — why is the user asking? where did this come from?>
tlp_ceiling: <optional — CLEAR/GREEN/AMBER/AMBER+STRICT/RED, default AMBER>

Procedure

1. Validate + triage (before spending API quota)

  • Reject RFC 1918 / loopback / link-local / documentation-range (192.0.2.0/24, 198.51.100.0/24, 203.0.113.0/24) unless the user specifically asks — these are not public infrastructure and enrichment is wasted
  • Check if the IP is already in data/iocs/ — if yes, load existing enrichment and note age
  • Decide whether to include Censys (skip by default due to 250/month quota)

2. Parallel lookups

Invoke in parallel (agents should batch these calls):

/lookup-liberty91 ioc-lookup <value>   # FIRST-PARTY, if $LIBERTY91_API_KEY is set — run first: is this already in your own platform, with what verdict, confidence and TLP?
/lookup-greynoise ip <value>    # FAST — filter out noise early
/lookup-virustotal ip <value>   # reputation aggregate
/lookup-abuseipdb ip <value>    # abuse reports
/lookup-shodan ip <value>       # infrastructure fingerprint
/lookup-otx ip <value>          # community pulses

Add to the same parallel batch if credentials are configured:

/lookup-reversinglabs ip <value>                 # if $REVERSINGLABS_USER is set. Network reputation from RL's malware-analysis corpus — files seen contacting this IP, RL classification. One API call, no pivot fan-out.
/lookup-crowdstrike indicator <value>            # if $CROWDSTRIKE_CLIENT_ID is set. Falcon Intel view — malicious confidence, linked actors + malware families, report refs. One API call.
/lookup-misp search-attributes --value <value>   # if $MISP_URL + $MISP_KEY are set. Internal correlation against your own catalogued events.
/lookup-opencti lookup <value>                   # if $OPENCTI_URL + $OPENCTI_TOKEN are set. Internal correlation against your OpenCTI knowledge base (observables + indicators in one call).

Optionally add (escalation — separate API budget):

/lookup-censys ip <value>                                                # ONLY if Shodan insufficient + quota available
/lookup-reversinglabs ip <value> --pivot files,domains,urls --max-results 25
# only when you want sample fan-out from the IP — each --pivot entry is a separate API call

3. Short-circuit on noise

If GreyNoise returns noise: true and classification: benign, AND AbuseIPDB confidence <10, this is likely mass-scanning background noise. Summarize briefly and stop — no need to burn full analysis on it. Note: noise: true with classification: malicious is still interesting (opportunistic attacker infrastructure).

4. Consolidate

Build a unified view:

indicator: <ip>
type: ipv4 | ipv6
consolidated_verdict: malicious | suspicious | clean | benign-noise | unknown
detection_signals:
  - source: virustotal
    detection_ratio: <X/Y>
    verdict: <...>
  - source: abuseipdb
    abuse_confidence: <0-100>
    recent_reports: <count>
  - source: greynoise
    noise: <bool>
    riot: <bool>
    classification: <...>
infrastructure:
  asn: <...>
  org: <...>
  country: <...>
  open_ports: <[...] from shodan>
  services: <top 5 from shodan>
  vulnerabilities: <CVE list>
  certificates: <present/absent>
community_context:
  pulse_count: <from otx>
  key_actors_attributed: <any pulse authors or malware families?>
pivot_candidates:
  - type: domain
    value: <hostname from shodan/vt>
    reason: <"resolves to this IP", "shares cert", etc.>
  - type: hash
    value: <communicating_files from vt>
    reason: <...>

5. Prioritize follow-up IOCs

Rank pivot candidates by investigative value:

  1. Certificate matches (highest — strong infrastructure link)
  2. Communicating files (medium-high — active C2 signal)
  3. Passive DNS resolutions (medium — historical association)
  4. Recent WHOIS/registration overlaps (medium)
  5. Generic ASN/org neighbors (low)

Return the top 3-5 candidates to the invoking skill or orchestrator for optional deep-dive via /domain-investigation, /hash-investigation, or /indicator-pivoting.

5b. Exposure check — was it seen in OUR environment? (if Sentinel is configured)

If $SENTINEL_WORKSPACE_ID + app credentials are set and the verdict is malicious or suspicious, chain /lookup-sentinel to sweep the IP across the workspace's available tables (DeviceNetworkEvents, CommonSecurityLog, SigninLogs, … — that skill discovers what exists before querying). Report hits with first/last seen and affected devices/accounts; report a miss as "not observed in collected telemetry over <window>", never "not compromised". Skip for benign-noise verdicts.

6. Apply rigor

Before returning, chain:

  • /source-assessment — rate each lookup result (defaults in each lookup skill's frontmatter)
  • /tlp-guide — mark the investigation output (default AMBER)
  • /confidence-levels — score the consolidated verdict (MISP 0-100)

7. Return

title: "IP Investigation: <ip>"
tlp: AMBER
confidence: <0-100>
summary: <2-3 sentences>
verdict: <...>
detection_signals: <...>
infrastructure: <...>
community_context: <...>
pivot_candidates: <top 3-5>
source_ratings:
  virustotal: B2
  shodan: B2
  ...
investigation_date: <ISO8601>
recommended_next_steps:
  - <e.g., "pivot on domain X via /domain-investigation">

When to stop

  • GreyNoise benign-noise short-circuit (step 3)
  • All lookups returned "not found" → report no signal; don't escalate
  • Clear malicious verdict with high confidence → return; leave deeper pivots to the user

Rate-limit awareness

Respect the lookup skills' own rate limits. Parallelizing 5 lookups on one IP = 5 API calls total, well within all free tiers. Don't loop over a list without throttling — use /ioc-enrichment-workflow for bulk.

Related skills

  • /domain-investigation, /hash-investigation, /url-investigation — sibling investigation skills
  • /indicator-pivoting — deeper graph traversal from a starting IOC
  • /ioc-enrichment-workflow — bulk IOC list enrichment
  • /threat-actor-profiling — if the IP is attributed to a known actor, consider profiling
  • /cti-orchestrator — typically the caller