Module: PWN::AI::Agent::Extrospection

Defined in:
lib/pwn/ai/agent/extrospection.rb

Overview

PWN::AI::Agent::Extrospection is the outward-facing counterpart to PWN::AI::Agent::Learning (introspection).

PRIMARY INTENT — on-demand external sensing

Quickly explore external resources when that produces a more informed answer. Call a sense tool only when the question needs it:

"weather in Tokyo"        → verify / watch / TransparentBrowser
"what's on 101.1 FM?"     → rf_tune(freq: "101.1") → RDS / observe(:rf)
"CVE for openssl 3.0?"    → intel(query:) / verify(claim:, kind: :cve)
"did the target change?"  → watch(url:) / snapshot(sections: [:web])

Secondary / optional — ambient host baseline

snapshot / drift / correlate can record cheap local posture so the agent can tell "I called the API wrong" from "the world moved" (kernel upgrade, dongle unplugged). This is NEVER the reason to launch GUI / JVM / heavy-REPL binaries — those are presence-only. auto_extrospect, when enabled, uses only side-effect-free sections.

INTROSPECTIVE (self)        EXTROSPECTIVE (world)
----------------------      -------------------------------------
Metrics.record              Extrospection.intel/verify/watch/rf_tune (sense)
Learning.note_outcome       Extrospection.observe         (fact)
Learning.reflect            Extrospection.snapshot/drift  (baseline)
Learning.stats              Extrospection.correlate       (self×world)

PromptBuilder re-injects Extrospection.to_context on every turn. Persistence: ~/.pwn/extrospection.json across REPL restarts.

Constant Summary collapse

EXTRO_FILE = 
File.join(Dir.home, '.pwn', 'extrospection.json')
MAX_OBSERVATIONS = 
500
SAFE_VERSION_BINS =

CLI tools that accept a cheap, non-interactive --version / -V.

%w[nmap curl git ruby python3 gcc openssl docker].freeze
PRESENCE_ONLY_BINS =

GUI / JVM / heavy REPL / interactive tools — presence-only. NEVER spawn these from auto-probe (Burp Suite splash, ZAP UI, msfconsole, GQRX).

%w[burpsuite zaproxy msfconsole gqrx sqlmap].freeze
RF_BINS = 
%w[rtl_sdr rtl_test rtl_433 hackrf_info gqrx dump1090 multimon-ng SoapySDRUtil].freeze
OSINT_BINS = 
%w[whois dig host curl jq].freeze
SERIAL_BINS = 
%w[minicom picocom screen cu].freeze
TELECOMM_BINS = 
%w[baresip asterisk linphone sngrep].freeze
PACKET_BINS = 
%w[tshark tcpdump tcpreplay dumpcap].freeze
VISION_BINS = 
%w[tesseract zbarimg qrencode convert identify].freeze
VOICE_BINS = 
%w[sox espeak-ng espeak festival whisper spd-say arecord aplay].freeze
PROBE_BINS = 
(
  SAFE_VERSION_BINS + PRESENCE_ONLY_BINS + RF_BINS +
  OSINT_BINS + SERIAL_BINS + TELECOMM_BINS + PACKET_BINS +
  VISION_BINS + VOICE_BINS
).uniq.freeze
AUTO_SECTIONS =

Cheap, side-effect-free sections used by auto_extrospect. toolchain / rf / web / osint / serial / telecomm / packet / vision / voice are on-demand only (sense tools or explicit snapshot).

%i[host repo env].freeze
WEB_SHOT_DIR = 
File.join(Dir.home, '.pwn', 'extrospection', 'web')
DEFAULT_WEB_ANCHORS = 
%w[
  https://services.nvd.nist.gov/rest/json/cves/2.0
  https://www.exploit-db.com/
  https://raw.githubusercontent.com/0dayinc/pwn/master/lib/pwn/version.rb
].freeze
DEFAULT_OSINT_FEEDS =

Public / free OSINT anchors (no key). Keys unlock richer feeds via PWN::Env. Expanded from https://github.com/public-api-lists/public-api-lists (Anti-Malware, Security, Geocoding, Government, Health, Open Data, Vehicle, …).

%i[
  ip geo dns whois rdap crtsh bgpview shodan hunter
  phone fcc_id patent person username github wayback
  otx urlhaus threatfox urlscan hackertarget openfda
  nominatim opencorporates courtlistener sec_edgar vital_records
  ipapi_is iplocate ipwhois
  abuseipdb virustotal greynoise
  certspotter epss cisa_kev
  nhtsa nppes federal_register uk_police callook
  mac_vendor universities microlink
  agify genderize nationalize
  haveibeenpwned securitytrails
].freeze

Class Method Summary collapse

Class Method Details

.authors ⇒ Object

Author(s)

0day Inc. support@0dayinc.com



3541
3542
3543
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 3541

public_class_method def self.authors
  "AUTHOR(S):\n  0day Inc. <support@0dayinc.com>\n"
end

.auto_extrospect(opts = {}) ⇒ Object

Supported Method Parameters

PWN::AI::Agent::Extrospection.auto_extrospect( session_id: 'optional - id of the just-completed session (for tagging)' )

Called by Learning.auto_introspect when PWN::Env[:agent][:auto_extrospect] is truthy. Captures a fresh snapshot and, if drift is non-trivial, records it as an observation and a PWN::Memory :env fact so the NEXT session's system prompt already knows the world moved. Never raises.



1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 1344

public_class_method def self.auto_extrospect(opts = {})
  sid = opts[:session_id]
  return unless auto_extrospect_enabled?

  # Ambient baseline only — never toolchain / rf / web (those spawn
  # hardware probes or GUI binaries and belong on the *sense* path).
  res   = snapshot(persist: true, sections: AUTO_SECTIONS)
  delta = res[:drift]
  moved = Array(delta[:changed]).length + Array(delta[:added]).length + Array(delta[:removed]).length
  return res if moved.zero?

  summary = summarise_drift(delta: delta)
  observe(source: 'auto_extrospect', category: :env, data: summary, tags: ['drift', sid.to_s].reject(&:empty?))
  if defined?(PWN::Memory)
    key = :"extro_drift_#{res[:snapshot][:fingerprint]}"
    PWN::Memory.remember(key: key, value: "ENV DRIFT #{res[:snapshot][:captured_at]}: #{summary[0, 260]}", category: :env)
  end
  res
rescue StandardError => e
  warn "[pwn-ai/extrospection] auto_extrospect swallowed: #{e.class}: #{e.message}"
  nil
end

.correlate(opts = {}) ⇒ Object

Supported Method Parameters

findings = PWN::AI::Agent::Extrospection.correlate( limit: 'optional - max findings returned (default 10)' )

THE join between introspection and extrospection. Cross-references:

* Metrics tools with success_rate < 50 %  vs  toolchain drift / missing bins
* Learning failures                       vs  host / net drift on the same day
* Observations tagged :vuln / :intel      vs  installed package versions

Emits human-readable, actionable findings the model can reason on.



1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 1175

public_class_method def self.correlate(opts = {})
  limit    = opts[:limit] || 10
  findings = []
  delta    = drift(live: false)
  snap     = load[:snapshot] || {}

  # 1) failing tools whose backing binary vanished / changed version
  if defined?(Metrics)
    Metrics.summary(limit: 50).each do |m|
      next unless m[:success_rate] < 0.5 && m[:calls] > 2

      bin = m[:name].to_s.split('_').first
      tc  = Array(delta[:changed]).find { |c| c[:path].to_s.include?("toolchain.#{bin}") }
      miss = snap.dig(:toolchain, bin.to_sym).to_s.empty?
      next unless tc || miss

      findings << { kind: :tool_env_mismatch, tool: m[:name], success_rate: m[:success_rate], evidence: tc || "binary '#{bin}' not found in PATH", advice: "Re-verify `which #{bin}` / reinstall before relying on #{m[:name]}." }
    end
  end

  # 2) introspective failures coinciding with extrospective drift
  if defined?(Learning)
    Learning.outcomes(limit: 30, success: false).each do |o|
      day = o[:timestamp].to_s[0, 10]
      hit = Array(delta[:changed]).find { |c| c[:after].to_s.include?(day) || c[:path].to_s.match?(/kernel|repo|net/) }
      findings << { kind: :failure_during_drift, task: o[:task], on: day, drift: hit, advice: 'Environment changed around this failure — re-test under current snapshot before trusting the negative result.' } if hit
    end
  end

  # 3) intel observations matching installed components
  pkgs = snap[:toolchain] || {}
  observations(category: 'intel', limit: 100).each do |ob|
    blob = ob[:data].to_s.downcase
    pkgs.each do |bin, ver|
      next if ver.to_s.empty?
      next unless blob.include?(bin.to_s.downcase)

      findings << { kind: :intel_matches_host, component: bin, installed: ver, intel: ob[:data], source: ob[:source], advice: "Review #{ob[:source]} advisory for #{bin} #{ver} on this host." }
    end
  end

  # 4) :rf observations vs missing SDR hardware / binaries
  rf = snap[:rf] || {}
  hw_present = %i[rtl_sdr hackrf flipper gqrx_sock].any? { |k| rf_present?(val: rf[k]) }
  observations(category: 'rf', limit: 50).each do |ob|
    miss = RF_BINS.select { |b| pkgs[b.to_sym].to_s.empty? }
    if !hw_present
      findings << { kind: :rf_no_hardware, observation: ob[:data], source: ob[:source], target: ob[:target], advice: 'RF observation recorded but no SDR hardware detected in snapshot — plug in RTL-SDR/HackRF/Flipper or start gqrx (`-r`) before trusting RF results.' }
    elsif rf[:gqrx_sock] == false && ob[:source].to_s == 'gqrx'
      findings << { kind: :rf_gqrx_down, observation: ob[:data], target: ob[:target], advice: 'gqrx remote-control socket (127.0.0.1:7356) is closed — start gqrx with remote control enabled before re-running the scan.' }
    elsif !miss.empty?
      findings << { kind: :rf_toolchain_gap, missing: miss, observation: ob[:data], advice: "SDR toolchain gap: install #{miss.join(', ')} to decode/act on this RF observation." }
    end
  end

  # 5) :web drift on a target the same day a Learning failure references it —
  #    "your exploit stopped working because the TARGET changed, not your approach"
  if defined?(Learning)
    observations(category: 'web', limit: 50).each do |ob|
      tgt  = ob[:target].to_s
      host = safe_host(url: tgt)
      Learning.outcomes(limit: 30, success: false).each do |o|
        next unless o[:task].to_s.include?(host) || o[:details].to_s.include?(host)

        findings << { kind: :target_web_drift, target: tgt, dom_sha: ob.dig(:data, :dom_sha), task: o[:task], advice: "Target #{host} DOM changed (#{ob[:timestamp]}) around this failure — re-recon before assuming your technique is wrong." }
      end
    end
  end

  # 6) extro_verify refutations whose claim substring appears in a PWN::Memory :fact -> stale memory
  if defined?(PWN::Memory)
    observations(category: 'web', tag: 'stale', limit: 50).each do |ob|
      findings << { kind: :stale_memory_fact, key: ob[:target], evidence: ob[:data], advice: "PWN::Memory[:#{ob[:target]}] failed browser re-verification — audit or memory_forget it before it poisons future prompts." }
    end
  end

  # 7) :intel observations whose source anchor is currently probe_web-unreachable -> downgrade
  web = snap[:web] || {}
  web.each do |host, fp|
    next unless fp.is_a?(Hash) && (fp[:reachable] == false || fp[:status].to_i >= 500)

    observations(category: 'intel', limit: 100).each do |ob|
      next unless ob[:source].to_s.include?(host.to_s) || ob[:data].to_s.include?(host.to_s)

      findings << { kind: :intel_source_unreachable, source: host, status: fp[:status], intel: ob[:data].to_s[0, 120], advice: "Feed anchor #{host} is currently unreachable (#{fp[:status] || 'down'}) — treat this :intel as stale until probe_web sees it 2xx again." }
    end
  end

  # 8) raw drift as low-priority findings when nothing else matched
  Array(delta[:added]).first(5).each   { |c| findings << { kind: :env_added,   detail: c } } if findings.empty?
  Array(delta[:removed]).first(5).each { |c| findings << { kind: :env_removed, detail: c } } if findings.empty?

  findings.uniq.first(limit)
end

.drift(opts = {}) ⇒ Object

Supported Method Parameters

delta = PWN::AI::Agent::Extrospection.drift( live: 'optional - probe host NOW and diff vs stored snapshot (default true). When false, diff stored :snapshot vs stored :previous.' )



244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 244

public_class_method def self.drift(opts = {})
  live  = if opts.key?(:live)
            opts[:live] ? true : false
          else
            true
          end
  store = load
  if live
    after  = snapshot(persist: false)[:snapshot]
    before = store[:snapshot] || {}
  else
    after  = store[:snapshot] || {}
    before = store[:previous] || {}
  end
  compute_drift(before: before, after: after)
end

.help ⇒ Object

Display Usage for this Module



3547
3548
3549
3550
3551
3552
3553
3554
3555
3556
3557
3558
3559
3560
3561
3562
3563
3564
3565
3566
3567
3568
3569
3570
3571
3572
3573
3574
3575
3576
3577
3578
3579
3580
3581
3582
3583
3584
3585
3586
3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 3547

public_class_method def self.help
  puts <<~USAGE
    USAGE:
      PWN::AI::Agent::Extrospection.snapshot                          # probe host, persist, return {snapshot:, drift:}
      PWN::AI::Agent::Extrospection.drift(live: true)                 # what changed vs last snapshot
      PWN::AI::Agent::Extrospection.observe(source: 'nmap', category: :recon, target: '10.0.0.5', data: '22/tcp open ssh 9.6')
      PWN::AI::Agent::Extrospection.observe(source: 'gqrx', category: :rf, target: '433.920MHz', data: 'peak -34.2 dBFS bw=200k')
      PWN::AI::Agent::Extrospection.observations(category: 'recon', target: '10.0.0.5')
      PWN::AI::Agent::Extrospection.intel(query: 'openssl 3.0', record: true)
      PWN::AI::Agent::Extrospection.correlate                         # introspection x extrospection findings
      PWN::AI::Agent::Extrospection.to_context                        # injected by PromptBuilder
      PWN::AI::Agent::Extrospection.stats
      PWN::AI::Agent::Extrospection.auto_extrospect(session_id: sid)  # called from Learning.auto_introspect
      PWN::AI::Agent::Extrospection.snapshot(sections: %i[web])        # opt-in browser probe of web_anchors
      PWN::AI::Agent::Extrospection.watch(url: 'https://target/api/version')
      PWN::AI::Agent::Extrospection.rf_tune(freq: '101.1')                 # tune GQRX + RDS → now_playing
      PWN::AI::Agent::Extrospection.osint(query: '+13125551212', kind: :phone)
      PWN::AI::Agent::Extrospection.osint(query: '2ABIP-ESP32', kind: :fcc_id)
      PWN::AI::Agent::Extrospection.osint(query: 'US10123456', kind: :patent)
      PWN::AI::Agent::Extrospection.osint(query: '1HGCM82633A004352', kind: :vin)
      PWN::AI::Agent::Extrospection.osint(query: '00:11:22:33:44:55', kind: :mac)
      PWN::AI::Agent::Extrospection.osint(query: 'W1AW', kind: :callsign)
      PWN::AI::Agent::Extrospection.osint(query: 'CVE-2021-44228', kind: :cve)
      PWN::AI::Agent::Extrospection.serial_sense(payload: "ATI\r")
      PWN::AI::Agent::Extrospection.telecomm(action: :status)
      PWN::AI::Agent::Extrospection.packet_sense(action: :capture, filter: 'tcp port 443', count: 10)
      PWN::AI::Agent::Extrospection.vision(file: '/tmp/shot.png', action: :ocr)
      PWN::AI::Agent::Extrospection.voice_sense(action: :tts, text: 'hello from pwn')
      PWN::AI::Agent::Extrospection.verify(claim: 'CVE-2026-12345 affects OpenSSL 3.2.1')
      PWN::AI::Agent::Extrospection.revalidate_memory                  # cron: GC stale PWN::Memory :fact entries
      PWN::AI::Agent::Extrospection.reset

      PRIMARY use = on-demand sensing (intel / verify / watch / rf_tune / osint /
      serial_sense / telecomm / packet_sense / vision / voice_sense / observe /
      rf / web / osint / serial / telecomm / packet / vision / voice).
      auto_extrospect is OPTIONAL ambient baseline (host/repo/env only — never
      launches burpsuite/zaproxy/msfconsole/gqrx). Prefer calling sense tools
      when a question needs the outside world, not after every turn.

      Enable end-of-run ambient baseline with:
        PWN::Env[:ai][:agent][:auto_extrospect] = true   # sections: AUTO_SECTIONS

      Configure browser-backed :web probe / verify / watch with:
        PWN::Env[:ai][:agent][:extrospection][:web] =
          { anchors: [...], proxy: 'tor', max_anchors: 8, per_page_timeout: 15, screenshot: false, allow_targets: false }

      Configure RF sense (rf_tune) with:
        PWN::Env[:ai][:agent][:extrospection][:rf] =
          { host: '127.0.0.1', port: 7356, settle_secs: 8, ttl: 300 }

      Configure new limbs:
        PWN::Env[:ai][:agent][:extrospection][:osint]    = { ttl: 86400, api_keys: { shodan: '…', hunter: '…', abuseipdb: '…', virustotal: '…', greynoise: '…', haveibeenpwned: '…', securitytrails: '…' } }
        PWN::Env[:ai][:agent][:extrospection][:serial]   = { block_dev: '/dev/ttyUSB0', baud: 115200, settle_secs: 1.5 }
        PWN::Env[:ai][:agent][:extrospection][:telecomm] = { host: '127.0.0.1', port: 8000 }
        PWN::Env[:ai][:agent][:extrospection][:packet]   = { iface: 'eth0' }
        PWN::Env[:ai][:agent][:extrospection][:vision]   = { lang: 'eng' }
        PWN::Env[:ai][:agent][:extrospection][:voice]    = { ttl: 3600 }

      #{self}.authors
  USAGE
end

.intel(opts = {}) ⇒ Object

Supported Method Parameters

hits = PWN::AI::Agent::Extrospection.intel( query: 'required - keyword / product / CVE id to search external threat-intel for', feeds: 'optional - Array subset of [:nvd, :circl, :exploitdb] (default all)', limit: 'optional - max results per feed (default 5)', record: 'optional - also persist each hit as an observation (default false)' )

Best-effort external lookups. Network / API failures degrade to an empty result for that feed rather than raising, so learning never breaks the primary loop when offline.



273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 273

public_class_method def self.intel(opts = {})
  query = opts[:query].to_s.strip
  raise 'ERROR: query is required' if query.empty?

  feeds  = Array(opts[:feeds]).map(&:to_sym)
  feeds  = %i[nvd circl exploitdb] if feeds.empty?
  limit  = opts[:limit] || 5
  record = opts[:record] ? true : false

  results = {}
  results[:nvd]       = intel_nvd(query: query, limit: limit)       if feeds.include?(:nvd)
  results[:circl]     = intel_circl(query: query, limit: limit)     if feeds.include?(:circl)
  results[:exploitdb] = intel_exploitdb(query: query, limit: limit) if feeds.include?(:exploitdb)

  if record
    results.each do |feed, hits|
      Array(hits).each do |h|
        observe(source: feed.to_s, category: :intel, data: h, tags: ['intel', query], target: h.is_a?(Hash) ? h[:id].to_s : nil)
      end
    end
  end
  { query: query, feeds: feeds, results: results, total: results.values.flatten.compact.length }
end

.load ⇒ Object

Supported Method Parameters

store = PWN::AI::Agent::Extrospection.load



97
98
99
100
101
102
103
104
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 97

public_class_method def self.load
  FileUtils.mkdir_p(File.dirname(EXTRO_FILE))
  return { snapshot: {}, previous: {}, observations: [], updated_at: nil } unless File.exist?(EXTRO_FILE)

  JSON.parse(File.read(EXTRO_FILE), symbolize_names: true)
rescue StandardError
  { snapshot: {}, previous: {}, observations: [], updated_at: nil }
end

.observations(opts = {}) ⇒ Object

Supported Method Parameters

rows = PWN::AI::Agent::Extrospection.observations( limit: 'optional - max entries newest-first (default 50)', source: 'optional - filter by source substring', category: 'optional - filter by category', target: 'optional - filter by target substring', tag: 'optional - filter by tag substring', fresh_only: 'optional - drop entries whose ttl has expired (default false)' )



216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 216

public_class_method def self.observations(opts = {})
  limit  = opts[:limit] || 50
  src    = opts[:source].to_s.downcase
  cat    = opts[:category].to_s.downcase
  tgt    = opts[:target].to_s.downcase
  tag    = opts[:tag].to_s.downcase
  fresh  = opts[:fresh_only] ? true : false
  now    = Time.now.utc

  rows = Array(load[:observations])
  rows = rows.reject do |o|
    next true if fresh && o[:ttl] && (Time.parse(o[:timestamp].to_s) + o[:ttl].to_i) < now

    (src.empty? ? false : !o[:source].to_s.downcase.include?(src)) ||
      (cat.empty? ? false : o[:category].to_s.downcase != cat) ||
      (tgt.empty? ? false : !o[:target].to_s.downcase.include?(tgt)) ||
      (tag.empty? ? false : Array(o[:tags]).none? { |t| t.to_s.downcase.include?(tag) })
  rescue StandardError
    false
  end
  rows.reverse.first(limit)
end

.observe(opts = {}) ⇒ Object

Supported Method Parameters

obs = PWN::AI::Agent::Extrospection.observe( source: 'required - where the observation came from (nmap, shodan, burp, cve, human, ...)', data: 'required - the observation payload (String or Hash)', category: 'optional - :recon, :vuln, :intel, :target, :network, :env, :rf, :web, :osint, :serial, :telecomm, :packet, :vision, :voice, :misc (default :misc)', target: 'optional - host/ip/url/asset the observation is about', tags: 'optional - Array of String labels', ttl: 'optional - seconds until this observation is considered stale (default nil = forever)' )

Records a fact about the OUTSIDE world (as opposed to Learning.note_outcome which records a fact about the agent's own behaviour). Observations are re-injected via .to_context so recon findings and threat-intel persist across sessions.



182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 182

public_class_method def self.observe(opts = {})
  source = opts[:source].to_s
  data   = opts[:data]
  raise 'ERROR: source is required' if source.strip.empty?
  raise 'ERROR: data is required'   if data.to_s.strip.empty?

  entry = {
    id: Digest::SHA256.hexdigest("#{source}-#{data}-#{Time.now.to_f}")[0, 12],
    source: source,
    category: (opts[:category] || :misc).to_s,
    target: opts[:target].to_s.empty? ? nil : opts[:target].to_s,
    data: data.is_a?(String) ? data[0, 4_000] : data,
    tags: Array(opts[:tags]).map(&:to_s),
    ttl: opts[:ttl]&.to_i,
    timestamp: Time.now.utc.iso8601
  }
  store = load
  store[:observations] ||= []
  store[:observations] << entry
  store[:observations].shift(store[:observations].length - MAX_OBSERVATIONS) if store[:observations].length > MAX_OBSERVATIONS
  save(store: store)
  entry
end

.osint(opts = {}) ⇒ Object

============================================================

OSINT sense organ

Supported Method Parameters

result = PWN::AI::Agent::Extrospection.osint( query: 'required - phone, IP, domain, email, person name, FCC ID, patent #, username, VIN, MAC, callsign, …', kind: 'optional - auto|:ip|:geo|:dns|:whois|:rdap|:crtsh|:bgp|:shodan|:hunter|:phone|:fcc_id|:patent|:person|:username|:github|:wayback|:url|:company|:cik|:openfda|:vital_records|:threat|:vin|:mac|:callsign|:npi|:cve (default :auto)', feeds: 'optional - Array subset of DEFAULT_OSINT_FEEDS (default: auto-selected from kind)', limit: 'optional - max hits per feed (default 5)', record: 'optional - also observe(category: :osint) (default true)', ttl: 'optional - observation TTL seconds (default 86400)', api_keys: 'optional - Hash of hunter:, abuseipdb:, virustotal:, greynoise:, haveibeenpwned:, securitytrails: overriding PWN::Env / ENV' )

Aggregates as many public / free OSINT APIs as possible (sourced in part from public-api-lists/public-api-lists), with optional keyed feeds (Shodan / Hunter / AbuseIPDB / VirusTotal / GreyNoise / HaveIBeenPwned / SecurityTrails) when keys exist in PWN::Env or ENV. Best-effort: any unreachable feed degrades to an error hash rather than raising. First-class kinds cover: reverse phone, person / missing-person, patent, FCC ID, VIN (NHTSA), MAC OUI, ham callsign (Callook), IP/ASN/BGP + threat reputation (ipapi.is / iplocate / ipwho.is / AbuseIPDB / GreyNoise), CT (crt.sh + Cert Spotter), whois/RDAP, GitHub, Wayback, OTX / URLHaus / ThreatFox / urlscan, HackerTarget, openFDA, NPPES NPI, Nominatim, OpenCorporates, CourtListener, SEC EDGAR, Federal Register, UK Police, EPSS + CISA KEV, Microlink unfurl, universities, name demographics (Agify/Genderize/Nationalize), and vital-records public plans.



642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 642

public_class_method def self.osint(opts = {})
  query = opts[:query].to_s.strip
  raise 'ERROR: query is required' if query.empty?

  kind   = (opts[:kind] || :auto).to_s.to_sym
  kind   = detect_osint_kind(query: query) if kind == :auto
  feeds  = Array(opts[:feeds]).map(&:to_sym)
  feeds  = osint_feeds_for(kind: kind) if feeds.empty?
  limit  = (opts[:limit] || 5).to_i
  record = opts.key?(:record) ? !opts[:record].nil? && opts[:record] != false : true
  ttl    = (opts[:ttl] || osint_config[:ttl] || 86_400).to_i
  keys   = osint_api_keys(override: opts[:api_keys])

  results = { ok: true, query: query, kind: kind, feeds: {}, captured_at: Time.now.utc.iso8601 }
  feeds.each do |feed|
    results[:feeds][feed] = osint_dispatch(feed: feed, query: query, kind: kind, limit: limit, keys: keys)
  rescue StandardError => e
    results[:feeds][feed] = { error: "#{e.class}: #{e.message.to_s[0, 160]}" }
  end

  results[:summary] = build_osint_summary(results: results)
  if record
    observe(
      source: 'osint',
      category: :osint,
      target: query,
      data: results[:summary],
      tags: (['osint', kind.to_s] + feeds.map(&:to_s)).uniq,
      ttl: ttl
    )
    results[:observed] = true
  else
    results[:observed] = false
  end
  results
end

.packet_sense(opts = {}) ⇒ Object

============================================================

Packet sense organ

Supported Method Parameters

result = PWN::AI::Agent::Extrospection.packet_sense( action: 'optional - :inventory|:capture|:summarize_pcap (default :inventory)', iface: 'optional - capture interface (default first non-lo or "any")', filter: 'optional - BPF filter (e.g. "tcp port 443")', count: 'optional - packets to capture (default 20, max 200)', timeout: 'optional - capture seconds (default 5, max 60)', path: 'optional - pcap path for :summarize_pcap', record: 'optional - observe(category: :packet) (default true)', ttl: 'optional - observation TTL (default 600)' )

Passive L2/L3 sense via tshark/tcpdump when present; pcap summarisation via PWN::Plugins::Packet + tshark. Capture is short and bounded so the agent never hangs mid-turn.



883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 883

public_class_method def self.packet_sense(opts = {})
  action  = (opts[:action] || :inventory).to_s.to_sym
  record  = opts.key?(:record) ? !opts[:record].nil? && opts[:record] != false : true
  ttl     = (opts[:ttl] || packet_config[:ttl] || 600).to_i
  iface   = (opts[:iface] || packet_config[:iface] || default_capture_iface).to_s
  filter  = opts[:filter].to_s
  count   = (opts[:count] || 20).to_i.clamp(1, 200)
  timeout = (opts[:timeout] || 5).to_i.clamp(1, 60)
  path    = opts[:path].to_s

  inv = probe_packet
  out = {
    ok: true,
    action: action,
    iface: iface,
    inventory: inv,
    captured_at: Time.now.utc.iso8601
  }

  case action
  when :inventory
    out[:ifaces] = inv[:ifaces]
    out[:bins]   = inv[:bins]
  when :capture
    tshark = sh(cmd: 'which tshark 2>/dev/null')
    tcpdump = sh(cmd: 'which tcpdump 2>/dev/null')
    if tshark.empty? && tcpdump.empty?
      out[:ok] = false
      out[:error] = 'neither tshark nor tcpdump found in PATH'
      out[:advice] = 'Install wireshark-common / tcpdump, or summarize an existing pcap with action: :summarize_pcap path:.'
    else
      pcap_out = File.join(Dir.home, '.pwn', 'extrospection', 'packet', "cap_#{Time.now.utc.strftime('%Y%m%d_%H%M%S')}.pcap")
      FileUtils.mkdir_p(File.dirname(pcap_out))
      if tshark.empty?
        bpf = filter.empty? ? '' : filter
        cmd = "timeout #{timeout + 2} tcpdump -i #{Shellwords.escape(iface)} -c #{count} -w #{Shellwords.escape(pcap_out)} #{Shellwords.escape(bpf)} 2>&1"
      else
        bpf = filter.empty? ? '' : "-f #{Shellwords.escape(filter)}"
        cmd = "timeout #{timeout + 2} tshark -i #{Shellwords.escape(iface)} -c #{count} -a duration:#{timeout} -w #{Shellwords.escape(pcap_out)} #{bpf} 2>&1"
      end
      log = sh(cmd: cmd)
      out[:pcap] = File.exist?(pcap_out) ? pcap_out : nil
      out[:log] = log.to_s[0, 500]
      out[:summary] = summarize_pcap_file(path: pcap_out) if out[:pcap]
      out[:ok] = !out[:pcap].nil?
      out[:error] = 'capture produced no pcap' unless out[:ok]
    end
  when :summarize_pcap
    if path.empty? || !File.exist?(path)
      out[:ok] = false
      out[:error] = "pcap not found: #{path.inspect}"
    else
      out[:path] = path
      out[:summary] = summarize_pcap_file(path: path)
    end
  else
    out[:ok] = false
    out[:error] = "unsupported action: #{action}"
  end

  if record
    blob = out[:summary].is_a?(Hash) ? out[:summary].to_json[0, 300] : out[:summary].to_s[0, 200]
    summary = "packet action=#{action} iface=#{iface} #{blob}"
    observe(source: 'packet', category: :packet, target: iface, data: summary, tags: ['packet', action.to_s], ttl: ttl)
    out[:observed] = true
    out[:obs_summary] = summary
  end
  out
end

.reset ⇒ Object

Supported Method Parameters

PWN::AI::Agent::Extrospection.reset



1370
1371
1372
1373
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 1370

public_class_method def self.reset
  FileUtils.rm_f(EXTRO_FILE)
  { cleared: true }
end

.revalidate_memory(opts = {}) ⇒ Object

Supported Method Parameters

report = PWN::AI::Agent::Extrospection.revalidate_memory( limit: 'optional - max :fact entries to check (default 25)', proxy: 'optional - upstream proxy for TransparentBrowser' )

The browser as garbage-collector for PWN::Memory. Walks :fact entries containing a CVE id, version string or URL, runs verify() on each and prefixes stale ones with [UNVERIFIED yyyy-mm-dd] so the injected MEMORY block stops calcifying into confidently-wrong priors. Designed to be scheduled: cron_create(schedule:'0 4 * * 0', ruby:'PWN::AI::Agent::Extrospection.revalidate_memory')



1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 1139

public_class_method def self.revalidate_memory(opts = {})
  return { checked: 0, refuted: [], confirmed: [], unknown: [] } unless defined?(PWN::Memory)

  lim   = opts[:limit] || 25
  proxy = opts[:proxy] || web_config[:proxy]
  rx    = %r{(CVE-\d{4}-\d{4,}|v?\d+\.\d+\.\d+|https?://\S+)}i
  out   = { checked: 0, refuted: [], confirmed: [], unknown: [] }

  PWN::Memory.load.each do |key, v|
    break if out[:checked] >= lim
    next unless v.is_a?(Hash) && v[:category].to_s == 'fact' && v[:value].to_s =~ rx
    next if v[:value].to_s.start_with?('[UNVERIFIED')

    res = verify(claim: v[:value].to_s[0, 400], commit: false, proxy: proxy)
    out[:checked] += 1
    out[res[:verdict]] << key
    next unless res[:verdict] == :refuted && res[:confidence] >= 0.6

    stamp = Time.now.utc.strftime('%Y-%m-%d')
    PWN::Memory.remember(key: key, value: "[UNVERIFIED #{stamp}] #{v[:value]}", category: :fact)
    observe(source: 'extro_verify', category: :web, target: key.to_s, data: "Memory :fact '#{key}' failed re-validation (#{res[:confidence]})", tags: %w[stale memory])
  end
  out
end

.rf_tune(opts = {}) ⇒ Object

Supported Method Parameters

result = PWN::AI::Agent::Extrospection.rf_tune( freq: 'required - frequency: "101.1", "101.1 FM", "101.100.000", 101_100_000, ...', host: 'optional - GQRX remote-control host (default 127.0.0.1)', port: 'optional - GQRX remote-control port (default 7356)', settle_secs: 'optional - seconds to sample RDS after tune (default 8)', rds: 'optional - force RDS on/off (default: auto when FM broadcast / band-plan decoder=:rds)', demodulator_mode: 'optional - e.g. :WFM_ST, :WFM, :FM, :AM (default from band-plan / FM range)', bandwidth: 'optional - passband Hz string, e.g. "200.000" (default from band-plan)', record: 'optional - also observe(category: :rf) so it hits EXTROSPECTION (default true)', ttl: 'optional - observation TTL seconds (default 300 — radio content is ephemeral)' )

RF sense organ — the RF analogue of extro_watch / extro_verify. Tunes a running GQRX instance (never launches the GUI), demodulates, measures strength, and when appropriate samples RDS (PI / PS / RadioText) so questions like "what's playing on 101.1?" have a live answer. Requires GQRX remote control already listening (default :7356) and an SDR attached; fails fast with actionable advice otherwise. On success, records observe(category: :rf, source: 'gqrx') so correlate and to_context keep the agent aware of what was last heard.



464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 464

public_class_method def self.rf_tune(opts = {})
  raw_freq = opts[:freq]
  raise 'ERROR: freq is required' if raw_freq.nil? || raw_freq.to_s.strip.empty?

  host        = (opts[:host] || rf_config[:host] || '127.0.0.1').to_s
  port        = (opts[:port] || rf_config[:port] || 7356).to_i
  settle      = (opts[:settle_secs] || rf_config[:settle_secs] || 8).to_f
  settle      = 1.0 if settle < 1.0
  settle      = 30.0 if settle > 30.0
  record      = opts.key?(:record) ? !opts[:record].nil? && opts[:record] != false : true
  ttl         = (opts[:ttl] || rf_config[:ttl] || 300).to_i
  force_rds   = opts.key?(:rds) ? opts[:rds] : nil

  hz_i, freq_label = normalize_rf_freq(freq: raw_freq)
  plan = match_rf_band_plan(hz: hz_i)
  demod = (opts[:demodulator_mode] || (plan && plan[:demodulator_mode]) || default_rf_demod(hz: hz_i)).to_s.upcase
  bandwidth = (opts[:bandwidth] || (plan && plan[:bandwidth]) || default_rf_bandwidth(hz: hz_i)).to_s
  passband_hz = begin
    require 'pwn/sdr' unless defined?(PWN::SDR)
    PWN::SDR.hz_to_i(freq: bandwidth)
  rescue StandardError
    bandwidth.to_s.gsub(/[^\d]/, '').to_i
  end
  passband_hz = 200_000 if passband_hz <= 0
  decoder_key = plan && plan[:decoder]
  band_name   = plan && plan[:name]
  do_rds = if force_rds.nil?
             decoder_key.to_s == 'rds' || hz_i.between?(87_500_000, 108_100_000)
           else
             !force_rds.nil? && force_rds != false
           end

  unless tcp_open?(host: host, port: port)
    advice = 'Start GQRX with remote control enabled (Tools → Remote Control, port 7356) and an SDR attached, then retry extro_rf_tune.'
    err = {
      ok: false,
      error: "gqrx remote control not reachable at #{host}:#{port}",
      advice: advice,
      freq: freq_label,
      hz: hz_i,
      band_plan: band_name,
      demodulator_mode: demod,
      bandwidth: bandwidth
    }
    observe(source: 'gqrx', category: :rf, target: freq_label, data: err, tags: %w[rf_tune unreachable], ttl: ttl) if record
    return err
  end

  require 'pwn/sdr/gqrx' unless defined?(PWN::SDR::GQRX)
  sock = nil
  begin
    sock = PWN::SDR::GQRX.connect(target: host, port: port)
    # Ensure DSP is running so strength/RDS update.
    begin
      dsp = PWN::SDR::GQRX.cmd(gqrx_sock: sock, cmd: 'u DSP').to_s.strip
      PWN::SDR::GQRX.cmd(gqrx_sock: sock, cmd: 'U DSP 1', resp_ok: 'RPRT 0') if dsp == '0'
    rescue StandardError
      nil
    end

    PWN::SDR::GQRX.cmd(
      gqrx_sock: sock,
      cmd: "M #{demod} #{passband_hz}",
      resp_ok: 'RPRT 0'
    )
    PWN::SDR::GQRX.cmd(
      gqrx_sock: sock,
      cmd: "F #{hz_i}",
      resp_ok: 'RPRT 0'
    )
    sleep 0.4

    strength = begin
      PWN::SDR::GQRX.cmd(gqrx_sock: sock, cmd: 'l STRENGTH').to_f
    rescue StandardError
      nil
    end
    mode_now = begin
      PWN::SDR::GQRX.cmd(gqrx_sock: sock, cmd: 'm').to_s.strip
    rescue StandardError
      "#{demod} #{passband_hz}"
    end
    tuned = begin
      PWN::SDR::GQRX.cmd(gqrx_sock: sock, cmd: 'f').to_s.strip
    rescue StandardError
      freq_label
    end

    rds = do_rds ? sample_rds(gqrx_sock: sock, settle_secs: settle) : nil

    payload = {
      ok: true,
      freq: freq_label,
      hz: hz_i,
      tuned: tuned,
      strength_dbfs: strength,
      demodulator_mode: demod,
      mode: mode_now,
      bandwidth: bandwidth,
      passband_hz: passband_hz,
      band_plan: band_name,
      decoder: decoder_key,
      rds: rds,
      now_playing: rds && (rds[:radiotext].to_s.strip.empty? ? nil : rds[:radiotext].to_s.strip),
      station: rds && (rds[:station].to_s.strip.empty? ? nil : rds[:station].to_s.strip),
      host: host,
      port: port,
      captured_at: Time.now.utc.iso8601
    }

    if record
      summary = build_rf_summary(payload: payload)
      observe(
        source: 'gqrx',
        category: :rf,
        target: freq_label,
        data: summary,
        tags: ['rf_tune', band_name, decoder_key, ('rds' if rds)].compact.map(&:to_s),
        ttl: ttl
      )
      payload[:observed] = true
      payload[:summary] = summary
    else
      payload[:observed] = false
    end

    payload
  rescue StandardError => e
    {
      ok: false,
      error: "#{e.class}: #{e.message}",
      freq: freq_label,
      hz: hz_i,
      band_plan: band_name,
      advice: 'Confirm GQRX is running with remote control, the SDR is not claimed by another process, and the frequency is in-band for the attached radio.'
    }
  ensure
    begin
      PWN::SDR::GQRX.cmd(gqrx_sock: sock, cmd: 'U RDS 0') if sock && do_rds
    rescue StandardError
      nil
    end
    begin
      PWN::SDR::GQRX.disconnect(gqrx_sock: sock) if sock
    rescue StandardError
      nil
    end
  end
end

.save(opts = {}) ⇒ Object

Supported Method Parameters

PWN::AI::Agent::Extrospection.save( store: 'required - Hash returned by .load / mutated in place' )



111
112
113
114
115
116
117
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 111

public_class_method def self.save(opts = {})
  store = opts[:store] ||= { snapshot: {}, previous: {}, observations: [] }
  store[:updated_at] = Time.now.utc.iso8601
  FileUtils.mkdir_p(File.dirname(EXTRO_FILE))
  File.write(EXTRO_FILE, JSON.pretty_generate(store))
  store
end

.serial_sense(opts = {}) ⇒ Object

============================================================

Serial sense organ

Supported Method Parameters

result = PWN::AI::Agent::Extrospection.serial_sense( block_dev: 'optional - device path (default first ttyUSB/ttyACM or config)', baud: 'optional - baud rate (default 9600)', payload: 'optional - String or byte Array to write (e.g. "ATI\r" or [0x41,0x54])', settle_secs:'optional - seconds to read after write (default 1.5)', data_bits: 'optional (default 8)', stop_bits: 'optional (default 1)', parity: 'optional :none|:even|:odd (default :none)', record: 'optional - observe(category: :serial) (default true)', ttl: 'optional - observation TTL (default 600)' )

Passive inventory via snapshot(sections:[:serial]); this verb is the active serial sense — open a device (PWN::Plugins::Serial), optional payload write, drain response, disconnect. Never keeps the port open across calls so other tools can claim the bus.



699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 699

public_class_method def self.serial_sense(opts = {})
  cfg = serial_config
  block_dev = (opts[:block_dev] || cfg[:block_dev] || first_serial_dev).to_s
  baud      = (opts[:baud] || cfg[:baud] || 9600).to_i
  settle    = (opts[:settle_secs] || cfg[:settle_secs] || 1.5).to_f
  settle    = 0.2 if settle < 0.2
  settle    = 30.0 if settle > 30.0
  record    = opts.key?(:record) ? !opts[:record].nil? && opts[:record] != false : true
  ttl       = (opts[:ttl] || cfg[:ttl] || 600).to_i
  payload   = opts[:payload]

  inv = probe_serial
  if block_dev.empty? || !File.exist?(block_dev)
    err = {
      ok: false,
      error: "serial device not found: #{block_dev.inspect}",
      advice: 'Plug a USB-UART / modem / Arduino or pass block_dev: "/dev/ttyUSB0". See probe_serial inventory.',
      inventory: inv
    }
    observe(source: 'serial', category: :serial, target: block_dev, data: err, tags: %w[serial_sense unreachable], ttl: ttl) if record
    return err
  end

  require 'pwn/plugins/serial' unless defined?(PWN::Plugins::Serial)
  serial_obj = nil
  begin
    serial_obj = PWN::Plugins::Serial.connect(
      block_dev: block_dev,
      baud: baud,
      data_bits: (opts[:data_bits] || 8).to_i,
      stop_bits: (opts[:stop_bits] || 1).to_i,
      parity: (opts[:parity] || :none).to_sym
    )
    PWN::Plugins::Serial.flush_session_data
    PWN::Plugins::Serial.request(serial_obj: serial_obj, payload: payload) if payload
    sleep settle
    raw = Array(PWN::Plugins::Serial.dump_session_data)
    text = raw.join.force_encoding('UTF-8').scrub
    hex  = raw.map do |b|
      case b
      when Integer then format('%02x', b & 0xff)
      when String  then b.bytes.map { |x| format('%02x', x) }.join
      else b.to_s.unpack1('H*').to_s
      end
    end.join(' ')
    line = begin
      PWN::Plugins::Serial.get_line_state(serial_obj: serial_obj)
    rescue StandardError
      nil
    end
    modem = begin
      PWN::Plugins::Serial.get_modem_params(serial_obj: serial_obj)
    rescue StandardError
      nil
    end

    payload_out = {
      ok: true,
      block_dev: block_dev,
      baud: baud,
      bytes: raw.length,
      text: text[0, 2_000],
      hex: hex[0, 2_000],
      line_state: line,
      modem_params: modem,
      inventory: inv,
      captured_at: Time.now.utc.iso8601
    }
    if record
      summary = "serial #{block_dev}@#{baud} bytes=#{raw.length} text=#{text.to_s.gsub(/\s+/, ' ')[0, 120]}"
      observe(source: 'serial', category: :serial, target: block_dev, data: summary, tags: %w[serial_sense], ttl: ttl)
      payload_out[:observed] = true
      payload_out[:summary] = summary
    end
    payload_out
  rescue StandardError => e
    {
      ok: false,
      error: "#{e.class}: #{e.message}",
      block_dev: block_dev,
      advice: 'Confirm device path, permissions (dialout group), and that no minicom/screen is holding the port.'
    }
  ensure
    begin
      PWN::Plugins::Serial.disconnect(serial_obj: serial_obj) if serial_obj
    rescue StandardError
      nil
    end
  end
end

.snapshot(opts = {}) ⇒ Object

Supported Method Parameters

snap = PWN::AI::Agent::Extrospection.snapshot( persist: 'optional - Boolean, write snapshot to disk & rotate previous (default true)', sections: 'optional - Array subset of [:host, :net, :toolchain, :repo, :env, :rf, :web, :osint, :serial, :telecomm, :packet, :vision, :voice] (default host/net/toolchain/repo/env/rf)' )

Ambient host baseline (secondary to sense tools like intel/verify/watch). :toolchain never spawns GUI/JVM tools (presence-only). auto_extrospect uses AUTO_SECTIONS (host/repo/env) only. When persist:true the prior snapshot is rotated into :previous so .drift can diff them.



130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 130

public_class_method def self.snapshot(opts = {})
  persist  = if opts.key?(:persist)
               opts[:persist] ? true : false
             else
               true
             end
  sections = Array(opts[:sections]).map(&:to_sym)
  sections = %i[host net toolchain repo env rf] if sections.empty?

  snap = {}
  snap[:host]      = probe_host      if sections.include?(:host)
  snap[:net]       = probe_net       if sections.include?(:net)
  snap[:toolchain] = probe_toolchain if sections.include?(:toolchain)
  snap[:repo]      = probe_repo      if sections.include?(:repo)
  snap[:env]       = probe_env       if sections.include?(:env)
  snap[:rf]        = probe_rf        if sections.include?(:rf)
  snap[:web]       = probe_web       if sections.include?(:web)
  snap[:osint]     = probe_osint     if sections.include?(:osint)
  snap[:serial]    = probe_serial    if sections.include?(:serial)
  snap[:telecomm]  = probe_telecomm  if sections.include?(:telecomm)
  snap[:packet]    = probe_packet    if sections.include?(:packet)
  snap[:vision]    = probe_vision    if sections.include?(:vision)
  snap[:voice]     = probe_voice     if sections.include?(:voice)
  snap[:captured_at] = Time.now.utc.iso8601
  snap[:fingerprint] = Digest::SHA256.hexdigest(JSON.generate(snap.except(:captured_at)))[0, 16]

  if persist
    store = load
    store[:previous] = store[:snapshot] || {}
    store[:snapshot] = snap
    save(store: store)
  end

  drift = compute_drift(before: load[:previous] || {}, after: snap)
  { snapshot: snap, drift: drift, persisted: persist }
end

.stats ⇒ Object

Supported Method Parameters

stats = PWN::AI::Agent::Extrospection.stats



1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 1309

public_class_method def self.stats
  store = load
  snap  = store[:snapshot] || {}
  delta = compute_drift(before: store[:previous] || {}, after: snap)
  {
    snapshot_captured_at: snap[:captured_at],
    snapshot_fingerprint: snap[:fingerprint],
    observations: Array(store[:observations]).length,
    drift_changed: Array(delta[:changed]).length,
    drift_added: Array(delta[:added]).length,
    drift_removed: Array(delta[:removed]).length,
    toolchain_bins: (snap[:toolchain] || {}).count { |_, v| !v.to_s.empty? },
    listening_ports: Array(snap.dig(:net, :listening)).length,
    rf_devices: (snap[:rf] || {}).values_at(:rtl_sdr, :hackrf, :flipper, :gqrx_sock).count { |v| rf_present?(val: v) },
    web_anchors: (snap[:web] || {}).count { |_, v| v.is_a?(Hash) && v[:reachable] },
    serial_devs: Array((snap[:serial] || {})[:devices]).length,
    packet_bins: ((snap[:packet] || {})[:bins] || {}).count { |_, v| !v.to_s.empty? },
    vision_bins: ((snap[:vision] || {})[:bins] || {}).count { |_, v| !v.to_s.empty? },
    voice_bins: ((snap[:voice] || {})[:bins] || {}).count { |_, v| !v.to_s.empty? },
    telecomm_http: (snap[:telecomm] || {})[:baresip_http] ? true : false,
    osint_feeds: Array((snap[:osint] || {})[:feeds_available]).length
  }
end

.telecomm(opts = {}) ⇒ Object

============================================================

Telecomm sense organ (SIP / VoIP / PSTN via BareSIP)

Supported Method Parameters

result = PWN::AI::Agent::Extrospection.telecomm( action: 'optional - :status|:dial|:hangup|:inventory (default :inventory)', target: 'optional - SIP URI / phone number for :dial (e.g. "sip:alice@example.com" or "+13125551212")', host: 'optional - BareSIP HTTP control host (default 127.0.0.1)', port: 'optional - BareSIP HTTP control port (default 8000)', record: 'optional - observe(category: :telecomm) (default true)', ttl: 'optional - observation TTL (default 600)' )

Telecomm analogue of rf_tune — senses live SIP / VoIP / PSTN state through a running BareSIP instance (never launches it). Status and inventory always; dial/hangup are explicit and OPSEC-sensitive.



806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 806

public_class_method def self.telecomm(opts = {})
  action = (opts[:action] || :inventory).to_s.to_sym
  host   = (opts[:host] || telecomm_config[:host] || '127.0.0.1').to_s
  port   = (opts[:port] || telecomm_config[:port] || 8000).to_i
  record = opts.key?(:record) ? !opts[:record].nil? && opts[:record] != false : true
  ttl    = (opts[:ttl] || telecomm_config[:ttl] || 600).to_i
  target = opts[:target].to_s.strip

  inv = probe_telecomm
  http_up = tcp_open?(host: host, port: port)

  out = {
    ok: true,
    action: action,
    host: host,
    port: port,
    baresip_http: http_up,
    inventory: inv,
    captured_at: Time.now.utc.iso8601
  }

  case action
  when :inventory, :status
    out[:status] = telecomm_baresip_cmd(host: host, port: port, cmd: '/?') if http_up
    out[:status] ||= inv
  when :dial
    if target.empty?
      out[:ok] = false
      out[:error] = 'target is required for action: :dial'
      out[:advice] = 'Pass target: "sip:user@host" or E.164 phone number.'
    elsif !http_up
      out[:ok] = false
      out[:error] = "baresip HTTP control not reachable at #{host}:#{port}"
      out[:advice] = 'Start baresip with HTTP module enabled (or PWN::Plugins::BareSIP.start) then retry.'
    else
      out[:dial] = telecomm_baresip_cmd(host: host, port: port, cmd: "/?dial=#{URI.encode_www_form_component(target)}")
    end
  when :hangup
    if http_up
      out[:hangup] = telecomm_baresip_cmd(host: host, port: port, cmd: '/?hangup')
    else
      out[:ok] = false
      out[:error] = "baresip HTTP control not reachable at #{host}:#{port}"
    end
  else
    out[:ok] = false
    out[:error] = "unsupported action: #{action}"
    out[:advice] = 'Use :inventory, :status, :dial, or :hangup.'
  end

  if record
    summary = "telecomm action=#{action} baresip_http=#{http_up} target=#{target.empty? ? '-' : target}"
    observe(source: 'telecomm', category: :telecomm, target: (target.empty? ? "#{host}:#{port}" : target), data: summary, tags: ['telecomm', action.to_s], ttl: ttl)
    out[:observed] = true
    out[:summary] = summary
  end
  out
end

.to_context(opts = {}) ⇒ Object

Supported Method Parameters

ctx = PWN::AI::Agent::Extrospection.to_context( drift_limit: 'optional - max drift lines (default 4)', obs_limit: 'optional - max observation lines (default 4)' )



1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 1276

public_class_method def self.to_context(opts = {})
  dlim = opts[:drift_limit] || 4
  olim = opts[:obs_limit]   || 4
  store = load
  return '' if (store[:snapshot] || {}).empty? && Array(store[:observations]).empty?

  snap  = store[:snapshot] || {}
  delta = compute_drift(before: store[:previous] || {}, after: snap)
  lines = []
  host  = snap[:host] || {}
  lines << "  host_fp   : #{snap[:fingerprint]}  (#{host[:os]} #{host[:kernel]}, #{host[:arch]})" if snap[:fingerprint]
  lines << "  captured  : #{snap[:captured_at]}" if snap[:captured_at]

  ch = Array(delta[:changed]).first(dlim).map { |c| "    ~ #{c[:path]}: #{c[:before].to_s[0, 40]} -> #{c[:after].to_s[0, 40]}" }
  ad = Array(delta[:added]).first(dlim).map   { |c| "    + #{c[:path]}: #{c[:after].to_s[0, 60]}" }
  rm = Array(delta[:removed]).first(dlim).map { |c| "    - #{c[:path]}: #{c[:before].to_s[0, 60]}" }
  unless ch.empty? && ad.empty? && rm.empty?
    lines << '  drift     :'
    lines.concat(ch).concat(ad).concat(rm)
  end

  obs = observations(limit: olim, fresh_only: true)
  unless obs.empty?
    lines << '  observed  :'
    obs.each { |o| lines << "    * [#{o[:category]}/#{o[:source]}] #{"#{o[:target]} — " if o[:target]}#{o[:data].to_s.gsub(/\s+/, ' ')[0, 120]}" }
  end

  "EXTROSPECTION (world-state; correlate with introspective failures)\n#{lines.join("\n")}\n\n"
end

.verify(opts = {}) ⇒ Object

Supported Method Parameters

verdict = PWN::AI::Agent::Extrospection.verify( claim: 'required - factual claim to fact-check against the live web', kind: 'optional - :cve | :version | :doc | :generic (default auto-detect)', url: 'optional - explicit URL the claim cites (forces kind: :doc)', commit: 'optional - write Mistakes/Memory/observe on refute/confirm (default true)', proxy: 'optional - upstream proxy for TransparentBrowser (e.g. "tor")' )

Browser-backed self fact-checking. Drives PWN::Plugins::TransparentBrowser in :headless mode against a canonical source for the claim class, renders the DOM (JS executed), and returns:

{ claim:, kind:, verdict: :confirmed|:refuted|:unknown,
confidence: 0.0..1.0,
evidence: [{url:, title:, excerpt:, dom_sha:, screenshot:}],
action_taken: :mistakes_record | :extro_observe | :learning_note | nil }

On :refuted -> Mistakes.record(tool:'assumption', error:claim) so the KNOWN MISTAKES block warns every future run. On :confirmed-> observe(category::intel, ttl:30d) reinforces w/ freshness. On :unknown -> Learning.note_outcome(success:false, tags:['needs_human']).

This is the extrospective mirror of Mistakes: a PROACTIVE trigger that catches the model being wrong about the world before a human does.



323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 323

public_class_method def self.verify(opts = {})
  claim = opts[:claim].to_s.strip
  raise 'ERROR: claim is required' if claim.empty?

  kind   = (opts[:kind] || detect_claim_kind(claim: claim, url: opts[:url])).to_sym
  commit = opts.key?(:commit) ? !opts[:commit].nil? && opts[:commit] != false : true
  proxy  = opts[:proxy] || web_config[:proxy]

  evidence = []
  verdict  = :unknown
  conf     = 0.0

  with_headless_browser(proxy: proxy) do |bo|
    case kind
    when :cve
      cve = claim[/CVE-\d{4}-\d{4,}/i].to_s.upcase
      %W[https://nvd.nist.gov/vuln/detail/#{cve} https://www.cve.org/CVERecord?id=#{cve}].each do |u|
        fp = fingerprint_page(browser_obj: bo, url: u)
        evidence << fp if fp
      end
      body = evidence.map { |e| e[:text].to_s }.join(' ')
      if evidence.empty? || evidence.all? { |e| e[:status].to_i >= 400 } || body =~ /could not be found|does not exist|reserved but/i
        verdict = :refuted
        conf = 0.8
      else
        prod = claim.sub(/CVE-\d{4}-\d{4,}/i, '').gsub(/affects?|is|in|the/i, ' ').split.select { |w| w.length > 2 }
        hit  = prod.count { |w| body.downcase.include?(w.downcase) }
        if prod.empty? || hit.to_f / prod.length >= 0.5
          verdict = :confirmed
          conf = prod.empty? ? 0.6 : [0.5 + (hit.to_f / prod.length * 0.5), 0.95].min
        else
          verdict = :refuted
          conf = 0.6
        end
      end
    when :doc
      u = opts[:url] || claim[URI::DEFAULT_PARSER.make_regexp(%w[http https])]
      fp = fingerprint_page(browser_obj: bo, url: u)
      evidence << fp if fp
      if fp.nil? || fp[:status].to_i >= 400 || fp[:reachable] == false
        verdict = :refuted
        conf = 0.7
      else
        snippet = claim.sub(u.to_s, '').strip
        overlap = fuzzy_overlap(needle: snippet, haystack: fp[:text].to_s)
        if fp[:title].to_s =~ /not found|404|error/i && overlap < 0.2
          verdict = :refuted
          conf = 0.7
        else
          verdict = overlap >= 0.4 ? :confirmed : :unknown
          conf    = overlap
        end
      end
    when :version
      proj = claim[/[A-Za-z][\w.+-]{2,}/].to_s
      ver  = claim[/v?\d+\.\d+(?:\.\d+)?/].to_s
      %W[https://rubygems.org/gems/#{proj} https://pypi.org/project/#{proj}/ https://github.com/search?q=#{URI.encode_www_form_component(proj)}&type=repositories].each do |u|
        fp = fingerprint_page(browser_obj: bo, url: u)
        evidence << fp if fp
      end
      body = evidence.map { |e| e[:text].to_s }.join(' ')
      if !ver.empty? && body.include?(ver)
        verdict = :confirmed
        conf = 0.7
      elsif body =~ /\d+\.\d+(?:\.\d+)?/
        verdict = :unknown
        conf = 0.3
      end
    else
      q  = URI.encode_www_form_component(claim)
      fp = fingerprint_page(browser_obj: bo, url: "https://html.duckduckgo.com/html/?q=#{q}")
      evidence << fp if fp
      overlap = fuzzy_overlap(needle: claim, haystack: fp ? fp[:text].to_s : '')
      verdict = overlap >= 0.5 ? :confirmed : :unknown
      conf    = overlap
    end
  end

  evidence.each { |e| e.delete(:text) }
  action = commit ? commit_verdict(claim: claim, kind: kind, verdict: verdict, confidence: conf, evidence: evidence) : nil
  { claim: claim, kind: kind, verdict: verdict, confidence: conf.round(2), evidence: evidence, action_taken: action }
rescue StandardError => e
  { claim: claim, kind: kind, verdict: :unknown, confidence: 0.0, evidence: evidence, error: "#{e.class}: #{e.message}", action_taken: nil }
end

.vision(opts = {}) ⇒ Object

============================================================

Vision / OCR sense organ

Supported Method Parameters

result = PWN::AI::Agent::Extrospection.vision( file: 'required - path to image / screenshot / PDF-page render', action: 'optional - :ocr|:barcode|:inventory (default :ocr when file given, else :inventory)', lang: 'optional - tesseract language (default eng)', record: 'optional - observe(category: :vision) (default true)', ttl: 'optional - observation TTL (default 86400)' )

Eyes on the host: OCR via PWN::Plugins::OCR (RTesseract / tesseract) and barcode/QR decode via zbarimg when present. Inventory only when no file is supplied.



968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 968

public_class_method def self.vision(opts = {})
  file   = opts[:file].to_s.strip
  action = (opts[:action] || (file.empty? ? :inventory : :ocr)).to_s.to_sym
  record = opts.key?(:record) ? !opts[:record].nil? && opts[:record] != false : true
  ttl    = (opts[:ttl] || vision_config[:ttl] || 86_400).to_i
  lang   = (opts[:lang] || vision_config[:lang] || 'eng').to_s

  inv = probe_vision
  out = {
    ok: true,
    action: action,
    inventory: inv,
    captured_at: Time.now.utc.iso8601
  }

  case action
  when :inventory
    # nothing else
  when :ocr
    if file.empty? || !File.exist?(file)
      out[:ok] = false
      out[:error] = "image not found: #{file.inspect}"
      out[:advice] = 'Pass file: "/path/to/image.png" (png/jpg/tiff/webp).'
    elsif inv.dig(:bins, :tesseract).to_s.empty? && !defined?(RTesseract)
      out[:ok] = false
      out[:error] = 'tesseract / rtesseract unavailable'
      out[:advice] = 'Install tesseract-ocr (+ eng language data) or the rtesseract gem.'
    else
      text = vision_ocr(file: file, lang: lang)
      out[:file] = file
      out[:text] = text.to_s[0, 8_000]
      out[:chars] = text.to_s.length
      out[:preview] = text.to_s.gsub(/\s+/, ' ')[0, 200]
    end
  when :barcode
    if file.empty? || !File.exist?(file)
      out[:ok] = false
      out[:error] = "image not found: #{file.inspect}"
    else
      codes = vision_barcodes(file: file)
      out[:file] = file
      out[:codes] = codes
    end
  else
    out[:ok] = false
    out[:error] = "unsupported action: #{action}"
  end

  if record
    summary = if out[:preview]
                "vision ocr file=#{File.basename(file)} chars=#{out[:chars]} preview=#{out[:preview]}"
              elsif out[:codes]
                "vision barcode file=#{File.basename(file)} codes=#{Array(out[:codes]).length}"
              else
                "vision action=#{action} tesseract=#{!inv.dig(:bins, :tesseract).to_s.empty?}"
              end
    observe(source: 'vision', category: :vision, target: (file.empty? ? 'inventory' : file), data: summary, tags: ['vision', action.to_s], ttl: ttl)
    out[:observed] = true
    out[:summary] = summary
  end
  out
end

.voice_sense(opts = {}) ⇒ Object

============================================================

Voice sense organ (TTS / STT / inventory)

Supported Method Parameters

result = PWN::AI::Agent::Extrospection.voice_sense( action: 'optional - :inventory|:tts|:stt (default :inventory)', text: 'optional - text to speak for :tts (or text_path:)', text_path: 'optional - path to text file for :tts', audio: 'optional - path to audio file for :stt', out: 'optional - output audio path for :tts (wav)', engine: 'optional - :espeak|:festival|:spd_say|:whisper (auto)', model: 'optional - whisper model (default tiny)', record: 'optional - observe(category: :voice) (default true)', ttl: 'optional - observation TTL (default 3600)' )

Wraps PWN::Plugins::Voice + system TTS/STT binaries (espeak-ng, festival, spd-say, whisper, sox). Inventory is always free; TTS/STT are on-demand and persist optional artefacts under ~/.pwn/extrospection/voice/.



1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 1051

public_class_method def self.voice_sense(opts = {})
  action = (opts[:action] || :inventory).to_s.to_sym
  record = opts.key?(:record) ? !opts[:record].nil? && opts[:record] != false : true
  ttl    = (opts[:ttl] || voice_config[:ttl] || 3_600).to_i
  inv    = probe_voice

  out = {
    ok: true,
    action: action,
    inventory: inv,
    captured_at: Time.now.utc.iso8601
  }

  case action
  when :inventory
    # nothing else
  when :tts
    text = opts[:text].to_s
    if text.empty? && opts[:text_path]
      text = begin
        File.read(opts[:text_path].to_s)
      rescue StandardError
        ''
      end
    end
    if text.strip.empty?
      out[:ok] = false
      out[:error] = 'text or text_path is required for action: :tts'
    else
      engine = (opts[:engine] || pick_tts_engine(inv: inv)).to_s.to_sym
      art = File.join(Dir.home, '.pwn', 'extrospection', 'voice')
      FileUtils.mkdir_p(art)
      wav = (opts[:out] || File.join(art, "tts_#{Time.now.utc.strftime('%Y%m%d_%H%M%S')}.wav")).to_s
      ok, log = voice_tts(text: text, engine: engine, out: wav, inv: inv)
      out[:engine] = engine
      out[:audio] = File.exist?(wav) ? wav : nil
      out[:log] = log.to_s[0, 400]
      out[:chars] = text.length
      out[:ok] = ok
      out[:error] = 'TTS engine failed' unless ok
    end
  when :stt
    audio = opts[:audio].to_s
    if audio.empty? || !File.exist?(audio)
      out[:ok] = false
      out[:error] = "audio not found: #{audio.inspect}"
      out[:advice] = 'Pass audio: "/path/to/recording.wav". Requires whisper binary or plugin path.'
    else
      engine = (opts[:engine] || pick_stt_engine(inv: inv)).to_s.to_sym
      text, log = voice_stt(audio: audio, engine: engine, model: opts[:model] || 'tiny', inv: inv)
      out[:engine] = engine
      out[:audio] = audio
      out[:text] = text.to_s[0, 8_000]
      out[:log] = log.to_s[0, 400]
      out[:ok] = !text.to_s.strip.empty?
      out[:error] = 'STT produced empty transcript' unless out[:ok]
    end
  else
    out[:ok] = false
    out[:error] = "unsupported action: #{action}"
  end

  if record
    summary = case action
              when :tts then "voice tts engine=#{out[:engine]} chars=#{out[:chars]} audio=#{out[:audio]}"
              when :stt then "voice stt engine=#{out[:engine]} text=#{out[:text].to_s.gsub(/\s+/, ' ')[0, 120]}"
              else "voice inventory espeak=#{!inv.dig(:bins, :'espeak-ng').to_s.empty? && inv.dig(:bins, :'espeak-ng') != ''} sox=#{!inv.dig(:bins, :sox).to_s.empty?}"
              end
    observe(source: 'voice', category: :voice, target: action.to_s, data: summary, tags: ['voice', action.to_s], ttl: ttl)
    out[:observed] = true
    out[:summary] = summary
  end
  out
end

.watch(opts = {}) ⇒ Object

Supported Method Parameters

obs = PWN::AI::Agent::Extrospection.watch( url: 'required - URL to render and fingerprint', selector: 'optional - CSS selector whose innerText to hash (default full body)', ttl: 'optional - seconds until stale (default 7 days)', proxy: 'optional - upstream proxy for TransparentBrowser' )

Passive change-detection on an external artefact you care about (target /api/version, vendor changelog, bounty scope page). Renders headless, hashes the DOM text, screenshots, and persists as observe(category: :web). On subsequent snapshot(sections:[:web]) or watch of the same URL, a hash mismatch surfaces in drift() as ~web..dom_sha exactly like ~toolchain.nmap today.



423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
 
# File 'lib/pwn/ai/agent/extrospection.rb', line 423

public_class_method def self.watch(opts = {})
  url = opts[:url].to_s
  raise 'ERROR: url is required' if url.empty?

  ttl   = opts[:ttl] || (7 * 24 * 3600)
  proxy = opts[:proxy] || web_config[:proxy]
  fp    = nil
  with_headless_browser(proxy: proxy) do |bo|
    fp = fingerprint_page(browser_obj: bo, url: url, selector: opts[:selector], screenshot: true)
  end
  raise "unreachable: #{url}" unless fp

  prior = observations(category: 'web', target: url, limit: 1).first
  changed = prior && prior.dig(:data, :dom_sha) && prior.dig(:data, :dom_sha) != fp[:dom_sha]
  data = fp.slice(:status, :final_url, :title, :dom_sha, :cert_fp, :cert_not_after, :screenshot).merge(excerpt: fp[:text].to_s[0, 200])
  observe(source: 'transparent_browser', category: :web, target: url, data: data, tags: (['watch'] + Array(opts[:tags])).compact, ttl: ttl)
  { url: url, changed: changed, prior_sha: prior && prior.dig(:data, :dom_sha), current: data }
end