Module: PWN::AI::Agent::Loop

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

Overview

The agent conversation loop:

build system prompt → call LLM with tools → if tool_calls: dispatch,
append role:'tool' results, loop → else: return text.

This replaces the regex-ReAct in PWN::Plugins::REPL :pwn_ai_hook with native function-calling. State (memory, skills, sessions) is all externalised — Loop.run is stateless aside from the messages array it builds.

NEGATIVE-FEEDBACK CLOSURE

Loop.run is where "learn from mistakes, don't repeat them" is actually enforced. On EVERY failed dispatch it:

1. Records the (tool, normalised_error) fingerprint into
 PWN::AI::Agent::Mistakes with a PERSISTENT cross-session count.
2. Reads that count back and, if it OR the in-turn count reaches
 REPEAT_THRESHOLD, prepends a hard "REPEATED FAILURE — change
 approach" guard to the tool result the model sees next.
3. Appends Mistakes.correction_hint (seen N×, sig, KNOWN FIX: …)
 so a previously-discovered fix is handed straight back to the
 model on the FIRST recurrence in a new session — it does not
 have to fail 3× again to re-learn what it already knew.

PromptBuilder.mistakes_block re-injects the top open mistakes and top known fixes into the system prompt of every future turn.

COMPLETION

The original request is the completion signal. TaskSummarizer and Policy are advisory (compass / rank). Loop keeps calling CORE_TOOLS until that request is done or truly blocked, then stops.

LOCAL-MODEL SCAFFOLDING

When the active engine is :ollama (or the corresponding :agent flags are set) Loop.run additionally:

* threads request → PromptBuilder for relevance-ranked MEMORY,
* threads request → Registry.definitions(relevance:) for a slimmed
tool set (:tool_router),
* splices Learning.exemplars_for(request:) between system and user
as few-shot behaviour retrieval,
* runs a plan-then-act pre-pass (:plan_first) so the model
externalises a tool plan before its first dispatch,
* escalates to a frontier persona for a 3-line corrective hint
once ≥ ESCALATE_AFTER_FAILS in-turn failures accumulate
(:escalation_persona) — the local model still produces the final
answer so Learning/Metrics stay attributed to :ollama.

Constant Summary collapse

DEFAULT_MAX_ITERS = 
777
ESCALATE_AFTER_FAILS = 
4
BOUNCE_FAIL_KEYS = 
%w[
  unsatisfied incomplete_final empty_final evidence_final
].freeze
ENGINE_MODS = 
{
  openai: 'PWN::AI::OpenAI',
  grok: 'PWN::AI::Grok',
  ollama: 'PWN::AI::Ollama',
  openwebui: 'PWN::AI::OpenWebUI',
  anthropic: 'PWN::AI::Anthropic',
  gemini: 'PWN::AI::Gemini'
}.freeze
HOT_WINDOW_SECS =

P17 — true when RECENT unresolved agent_loop / assistant_answer budget fingerprints dominate Mistakes.top. Sliding window + auto-cool so the loop's own exhaust-path Mistakes.record cannot permanently latch hot (scar 8ec3303ed69e self-latch). Do NOT deepen caps; cool the detector.

48 * 3600
HOT_COOL_MAX_RECENT =

<=1 budget hit in window => cooled (not hot)

1
PARK_COOL_SECS =

P17 rate-based cool/park for permanent budget scars (8ec3303ed69e). Leaves scar open but parks it so it stops dominating Mistakes.top. Resolve is rate-based only (external / after multi-day cool) — never because a guard patch landed. PARK_COOL_SECS (24h) is intentionally shorter than HOT_WINDOW (48h): once hot?=false, a single cooled scar must not keep owning Mistakes.top for another full day.

24 * 3600
PRIVILEGED_TOOLSETS = 
%w[cron swarm].freeze
SNAPSHOT_STALE_SECS = 
6 * 3600
INCOMPLETE_FINAL_RX =

P28 — incomplete / handoff finals: model emitted text-only before the goal was done ("shall I proceed?", "next step:", "want me to…"). Loop.run treats no-tool_calls as FINAL; this detector lets us refuse that handoff and keep the tool loop alive for multi-step autonomy.

Local/thinking models (gemma/Qwen abliterated etc.) often emit a monologue that NARRATES the next tool ("Wait, let's try hping3…") without producing native tool_calls or shell(...). Treat that as incomplete too so the loop re-pressures tools instead of FINAL.

/
  \b(shall\s+i|should\s+i|may\s+i|can\s+i|want\s+me\s+to|do\s+you\s+want\s+me|
     next\s+single\s+step|next\s+step\s*:|awaiting\s+your\s+(ok|approval|go-ahead|confirmation)|
     if\s+you(?:'d|\s+would)\s+like\s+me\s+to|say\s+the\s+word|confirm\s+(before|and\s+i)|
     ready\s+to\s+proceed|ok\s+to\s+(proceed|continue|apply)|proceed\?|
     continue\?|before\s+i\s+(apply|change|run|continue|proceed)|
     once\s+you\s+(confirm|approve)|let\s+me\s+know\s+if|
     i(?:'ll|\s+will)\s+wait\b|waiting\s+for\s+(your\s+)?(go|ok|approval|confirmation)
  )\b
/ix
MONOLOGUE_TOOL_INTENT_RX =

Narrated-intent monologue without a structured tool call. Distinct from INCOMPLETE_FINAL_RX (polite handoff to the human).

/
  \b(
    wait[,\s]+let'?s\s+try|
    let'?s\s+try\s+(one|to|again|hping|nmap|ping|sudo|shell|running|checking)|
    i\s+(?:will|'ll)\s+(?:just\s+)?(?:try|run|check|probe|scan|use)\b|
    actually,?\s+i\s+will\b|
    one\s+more\s+thing\b|
    if\s+it\s+fails\b.{0,80}\bthen\s+we\s+can\b|
    verification\s+complete\b|
    report\s+that\s+(?:the\s+)?verification\s+failed\b
  )
/ix
ACT_REQUEST_RX = 
/
  \b(write|create|implement|fix|patch|replace|refactor|overwrite|
     add (?:a |the )?|update|install|delete|remove|rename|
     regenerate|rebuild|document)
  \b
/ix
FORCED_WRAP_RX = 
/
  forced\s+to\s+a\s+final|were\s+not\s+written|not\s+written\s+to\s+disk|
  resume\s+from\s+the\s+table|remaining\s+block|
  were\s+not\s+applied|not\s+applied\s+in\s+this\s+turn|
  (?:do\s+that\s+)?next\s+time
/ix
LOOKUP_REQUEST_RX = 
/
  \b(what\s+is\s+my|hostname|uname|cwd|whoami|status|version|how\s+many)\b
/ix
HOST_PATH_RX =

Real filesystem paths only — not https://host.tld (that was matching //host.tld).

%r{(?:(?<![.:/])/(?!/)|\./)[\w./-]+\.\w+}
BROWSER_REQUEST_RX = 
/
  TransparentBrowser|browser_obj|\bdevtools\b|
  \b(navigate|dump_links|headless_?chrome|watir)\b
/ix
HOWTO_RX =

Request intent for routing (how-to vs act/recon vs pure recall/greeting). Local models thrash when pure explanation/recall/greeting asks are force-planned into multi-step host probes or multi-tool session archaeology. :howto → answer with explanation only (no plan_first / no live recon). :recall → prior-turn / vague memory cue; cheap path only. :greeting → short hello / light smalltalk; deterministic ack, no tools. :recon_act → live discovery (same tool loop as :act; no auth gate). :act → general agent work with tools.

/
  \b(
    how\s+to|how\s+do\s+i|how\s+can\s+i|how\s+would\s+i|how\s+does\s+one|
    what\s+is\s+the\s+(?:syntax|command|usage|flag|option)|
    explain\s+how|show\s+me\s+how|examples?\s+of\s+using|
    manual\s+for|usage\s+of|syntax\s+for|man\s+page
  )\b
/ix
RECALL_RX =

Pure prior-turn recall — must never enter plan_first / multi-tool loops. Covers both "what did I just say?" (user) and "how did you respond?" / "what did you just say?" (assistant) so last-turn injection is used.

/
  \A\s*(
    what\s+did\s+i\s+(just\s+)?say\??|
    what\s+did\s+i\s+(just\s+)?(?:ask|type|write|request)\??|
    what\s+was\s+my\s+last\s+(?:request|message|question|prompt|turn)\??|
    what\s+was\s+(?:the\s+)?(?:previous|prior|last)\s+(?:thing\s+i\s+said|request|message|turn)\??|
    remind\s+me\s+what\s+i\s+(?:just\s+)?(?:said|asked)\??|
    repeat\s+(?:my\s+)?(?:last|previous)\s+(?:request|message)\??|
    say\s+that\s+again\??|
    recollection\s+test\??|
    memory\s+recall\s+test\??|
    how\s+did\s+you\s+respond(?:\s+to\s+what\s+i\s+(?:just\s+)?(?:said|asked))?\??|
    how\s+did\s+you\s+(?:just\s+)?(?:answer|reply)(?:\s+to\s+(?:me|that|my\s+last))?\??|
    what\s+(?:was|is)\s+your\s+(?:last|previous|prior)\s+(?:answer|response|reply)\??|
    what\s+did\s+you\s+(?:just\s+)?(?:say|answer|reply|respond)\??|
    remind\s+me\s+what\s+you\s+(?:just\s+)?(?:said|answered|replied)\??|
    repeat\s+your\s+(?:last|previous)\s+(?:answer|response|reply)\??
  )\s*\z
/ix
VAGUE_MEMORY_RX =

Broader "use your memory / prior context" cues. Still cheap: inject last turn + at most one memory_recall; never multi-step plans.

/
  \b(
    what\s+did\s+i\s+(just\s+)?(?:say|ask|type|request)|
    what\s+was\s+my\s+last|
    how\s+did\s+you\s+respond|
    what\s+did\s+you\s+(?:just\s+)?(?:say|answer|reply|respond)|
    what\s+(?:was|is)\s+your\s+(?:last|previous|prior)\s+(?:answer|response|reply)|
    (?:without\s+looking\s+up).{0,40}(?:session|discussing|talking)|
    (?:from\s+)?(?:memory|context|earlier|previously|prior\s+turn)|
    (?:do\s+you\s+)?remember\s+what\s+(?:i|you)|
    recall\s+(?:what|my|your|the\s+last)|
    last\s+thing\s+(?:i|you)\s+said
  )\b
/ix
LAST_SESSION_RX = 
/\b(?:in|from|of)\s+(?:the\s+)?(?:last|previous|prior)\s+session\b|\blast\s+session\b/i
GREETING_RX =

Pure greeting / light smalltalk — never full :act tool loop. Anchored short forms only so "hi, please scan X" stays :act/:recon_act. Do NOT echo weather or invent social filler; answer_greeting is fixed.

/
  \A\s*(
    (?:hi|hello|howdy|hey|yo|sup|hiya|greetings)(?:\s*[.!?]*)?
    (?:\s*,?\s*(?:there|all|folks|team|everyone|y'?all))?
    |
    good\s+(?:morning|afternoon|evening|day|night)(?:\s*[.!?]*)?
    |
    (?:hi|hello|howdy|hey)(?:\s*[.!?*,]*)?\s+
    (?:it'?s|its|it\s+is)\s+
    (?:cloudy|sunny|rainy|raining|foggy|windy|stormy|nice|cold|hot|warm|
       beautiful|gloomy|overcast|clear|chilly|humid|snow(?:ing|y)?)
    (?:\s+out(?:\s+there)?)?(?:\s*[.!?]*)?
    |
    (?:hi|hello|howdy|hey)(?:\s*[.!?*,]*)?\s+
    (?:the\s+weather\s+is\s+\w+|what'?s\s+up|how\s+are\s+you|
       how'?s\s+it\s+going|how\s+goes\s+it)
    (?:\s*[.!?]*)?
  )\s*\z
/ix
LIVE_RECON_RX = 
/
  \b(
    (?:find|discover|enumerate|scan|sweep|probe|map)\s+
    (?:live\s+)?(?:hosts?|ips?|targets?|subnet|network|range)|
    live\s+hosts?\s+(?:can\s+you\s+)?find|
    what\s+live\s+hosts|
    ping\s+sweep\s+(?:of\s+)?(?:this|the|my)\s+
    |(?:run|do|perform)\s+(?:a\s+)?(?:ping\s+)?sweep
    |scan\s+(?:this|the|my)\s+(?:subnet|network|lan|range)
  )\b
/ix

Class Method Summary collapse

Class Method Details

.authors ⇒ Object

Author(s)

0day Inc. support@0dayinc.com



2471
2472
2473
 
# File 'lib/pwn/ai/agent/loop.rb', line 2471

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

.debug_on?(opts = {}) ⇒ Boolean

Returns:

  • (Boolean) 


65
66
67
68
69
70
71
72
73
74
 
# File 'lib/pwn/ai/agent/loop.rb', line 65

public_class_method def self.debug_on?(opts = {})
  return true if opts[:debug]
  return true if defined?(PWN::Plugins::Log) && PWN::Plugins::Log.debug_enabled?

  pry_on = defined?(Pry) && Pry.respond_to?(:config) &&
           Pry.config.respond_to?(:pwn_ai_debug) && Pry.config.pwn_ai_debug
  return true if pry_on

  false
end

.help ⇒ Object

Display Usage for this Module



2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
 
# File 'lib/pwn/ai/agent/loop.rb', line 2477

public_class_method def self.help
  puts <<~USAGE
    USAGE:
      final = PWN::AI::Agent::Loop.run(
        request: 'what does `id` return on this host?',
        session_id: PWN::Sessions.create[:id],
        enabled_toolsets: %w[terminal pwn memory skills],
        on_tool: ->(name, args, result) { puts "→ \#{name}: \#{result[0,1_024]}" },
        system_role_content: 'You are a helpful assistant that can call tools to answer questions.'
      )
      # Live task summaries (default ON): BEFORE each tool *collection*,
      # on_tool('task', high_level_brief, '') — one-to-many with real tools.
      # Task lines never carry a result payload (no result row in the TUI).
      # so repl.rb prints name=task with arg_preview=summary. Also coalesce bursts into
      # via on_tool only: [ ts → pwn-ai → task ] <brief> (no [pwn-ai/task] prefix)
      # Toggle via PWN::Env[:ai][:agent]:
      #   task_summary: true|false
      #   task_summary_every: 5          # emit every N tools
      #   task_summary_interval_s: 8.0   # or every N seconds
      #   task_summary_verbose: false

      Supported engines: #{ENGINE_MODS.keys.join(', ')}
      Set PWN::Env[:ai][:active] to choose; PWN::Env[:ai][:agent][:max_iters] to bound.

      Intent routing (all engines; critical for ollama/openwebui):
        how-to / usage questions → text-only explanation (no tools, no plan_first)
        pure prior-turn recall ("what did I just say?") → answer_recall (no tools)
        pure greeting / light smalltalk → answer_greeting (no tools, no weather echo)
        live sweeps are host-work goals like any other — pwn-ai does not
        decide authorization
      Local-model scaffolding (PWN::Env[:ai][:agent]):
        :plan_first          - Boolean, plan-then-act pre-pass (default: local engine :ollama/:openwebui)
        :tool_router         - Boolean/nil, slim Registry.definitions (nil=auto on for ollama)
        :escalation_persona  - Swarm persona name for frontier corrective hints when stuck
        :critic              - S3 constitutional critic before every final (Boolean)
        :red_team_plan       - S4 adversarial plan review after plan_first (Boolean)
        :counterfactual      - S2 A/B branch on REPEAT_THRESHOLD → DPO pair (Boolean)
        :hindsight           - C3 HER-relabel failures (Boolean, default true)
        :policy              - R5 live tabular Q / REINFORCE (Boolean, default true; advisory only)
        :verify_as_reward    - E3 ground every final via extro_verify (Boolean)

      P28 autonomy: incomplete-final detector refuses mid-goal handoffs.
      max_iters is Env or DEFAULT_MAX_ITERS (#{DEFAULT_MAX_ITERS});
      budget-hot / overconf do not shrink this request's runway.

      #{self}.authors
  USAGE
end

.needs_host_work?(opts = {}) ⇒ Boolean

Returns:

  • (Boolean) 


397
398
399
400
401
402
403
404
405
 
# File 'lib/pwn/ai/agent/loop.rb', line 397

public_class_method def self.needs_host_work?(opts = {})
  request = opts[:request].to_s
  return false if request.strip.empty?
  return false if world_knowledge?(request: request)

  true
rescue StandardError
  false
end

.ollama_wire_messages(opts = {}) ⇒ Object

Supported Method Parameters

wire = PWN::AI::Agent::Loop.ollama_wire_messages( messages: 'required - in-memory OpenAI-ish messages (may have String args)' )

Returns a deep-copied array safe for Ollama / Open WebUI ollama/api/chat:

  • parses JSON-string function.arguments into Hash/Array objects
  • coerces nil assistant content to '' when tool_calls present (Open WebUI GenerateChatCompletionForm rejects content:null alone)
  • drops _native_content / _text_tool_coerced / thinking private keys
  • stringifies Hash/Array message content (tool results) to JSON text


953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
 
# File 'lib/pwn/ai/agent/loop.rb', line 953

public_class_method def self.ollama_wire_messages(opts = {})
  messages = opts[:messages]
  Array(messages).filter_map do |m|
    next unless m.is_a?(Hash)

    role = (m[:role] || m['role']).to_s
    out = { role: role }

    tcs = m[:tool_calls] || m['tool_calls']
    wired_tcs = nil
    if tcs
      wired_tcs = Array(tcs).filter_map { |tc| ollama_wire_tool_call(tool_call: tc) }
      out[:tool_calls] = wired_tcs unless wired_tcs.empty?
    end

    if m.key?(:content) || m.key?('content')
      content = m.key?(:content) ? m[:content] : m['content']
      out[:content] = case content
                      when nil
                        # Open WebUI: null content without tool_calls 400s;
                        # with tool_calls prefer "" over null.
                        wired_tcs && !wired_tcs.empty? ? '' : nil
                      when String then content
                      when Hash, Array then JSON.generate(content)
                      else content.to_s
                      end
    elsif wired_tcs && !wired_tcs.empty?
      out[:content] = ''
    end

    name = m[:name] || m['name']
    out[:name] = name.to_s if name && !name.to_s.empty?

    tcid = m[:tool_call_id] || m['tool_call_id']
    out[:tool_call_id] = tcid.to_s if tcid && !tcid.to_s.empty?

    out
  end
end

.openai_wire_messages(opts = {}) ⇒ Object

Supported Method Parameters

wire = PWN::AI::Agent::Loop.openai_wire_messages( messages: 'required - in-memory OpenAI-ish messages (may have Hash args / internal keys)' )

Returns a deep-copied array safe for OpenAI / xAI chat.completions:

  • drops _native_content / _text_tool_coerced / thinking private keys
  • stringifies function.arguments maps
  • coerces Hash/non-string content to JSON/string (nil kept for assistant tool turns)


1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
 
# File 'lib/pwn/ai/agent/loop.rb', line 1037

public_class_method def self.openai_wire_messages(opts = {})
  messages = opts[:messages]
  Array(messages).filter_map do |m|
    next unless m.is_a?(Hash)

    role = (m[:role] || m['role']).to_s
    out = { role: role }

    if m.key?(:content) || m.key?('content')
      content = m.key?(:content) ? m[:content] : m['content']
      out[:content] = case content
                      when nil then nil
                      when String then content
                      when Hash, Array then JSON.generate(content)
                      else content.to_s
                      end
    end

    name = m[:name] || m['name']
    out[:name] = name.to_s if name && !name.to_s.empty?

    tcid = m[:tool_call_id] || m['tool_call_id']
    out[:tool_call_id] = tcid.to_s if tcid && !tcid.to_s.empty?

    tcs = m[:tool_calls] || m['tool_calls']
    if tcs
      wired = Array(tcs).filter_map { |tc| openai_wire_tool_call(tool_call: tc) }
      out[:tool_calls] = wired unless wired.empty?
    end

    out
  end
end

.request_intent(opts = {}) ⇒ Object 



1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
 
# File 'lib/pwn/ai/agent/loop.rb', line 1459

public_class_method def self.request_intent(opts = {})
  req = opts[:request].to_s
  return :empty if req.strip.empty?

  # Pure greeting / weather smalltalk before how-to/recon/act.
  # Deterministic short-circuit — never freeform model weather echo.
  return :greeting if req.match?(GREETING_RX)

  # Pure prior-turn recall before how-to/recon (short, decisive).
  return :recall if req.match?(RECALL_RX)

  # Vague memory cues that are still "about the prior turn" and not
  # general work ("remember what we decided about nmap and implement it"
  # stays :act because it pairs memory with a doing verb outside the cue).
  if req.match?(VAGUE_MEMORY_RX) && !req.match?(HOWTO_RX) && !req.match?(LIVE_RECON_RX)
    doing = req.match?(
      /\b(implement|fix|patch|refactor|run|execute|scan|write|edit|
          change|deploy|install|build|compile|commit|push)\b/ix
    )
    return :recall unless doing
  end

  if req.match?(LAST_SESSION_RX) && !req.match?(HOWTO_RX) && !req.match?(LIVE_RECON_RX)
    doing = req.match?(
      /\b(implement|fix|patch|refactor|run|execute|scan|write|edit|
          change|deploy|install|build|compile|commit|push)\b/ix
    )
    return :recall unless doing
  end

  # Live-action recon takes precedence over bare "how to" when both appear
  # only if the user clearly asks the agent to do the sweep here.
  live = req.match?(LIVE_RECON_RX) && req.match?(
    /\b(can\s+you|could\s+you|please|go\s+ahead|now|on\s+this\s+host|
        this\s+subnet|this\s+network|find\s+(?:for\s+me|me)|discover)\b/ix
  )
  return :recon_act if live || (req.match?(LIVE_RECON_RX) && !req.match?(HOWTO_RX))
  return :howto if req.match?(HOWTO_RX)
  # Interrogative documentation without "how to"
  if req.match?(/\b(what\s+(?:flags?|options?|switches?)|usage|syntax)\b/i) &&
     !req.match?(/\b(run|execute|scan|find|discover)\b/i)
    return :howto
  end

  :act
rescue StandardError
  :act
end

.run(opts = {}) ⇒ Object

Supported Method Parameters

final = PWN::AI::Agent::Loop.run( request: 'required - what the human typed', session_id: 'optional - PWN::Sessions id (transcript is appended to it)', enabled_toolsets: 'optional - subset of Registry.toolsets, or nil for all', on_tool: 'optional - ->(name, args, result) callback for live UI', system_role_content: 'optional - override default system prompt (built from session_id if not provided)' )



2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
 
# File 'lib/pwn/ai/agent/loop.rb', line 2032

public_class_method def self.run(opts = {})
  request = opts[:request].to_s
  session_id = opts[:session_id]
  on_tool = opts[:on_tool]
  start_debug_session(opts)
  loud_debug_tui!(debug: opts[:debug])
  debug_progress(msg: "Loop.run start request=#{request[0, 240]}", debug: opts[:debug])
  TurnFinalizer.enter_user_path! if defined?(TurnFinalizer)
  engine = active_engine
  local  = local_engine?(engine: engine)

  # Cheap intent/kind FIRST - before PromptBuilder / Registry / TaskSummarizer
  # so greetings, FYIs, how-tos, recall, and simple Qs never pay the fat path.
  intent = request_intent(request: request)
  if defined?(OpenGoal) && OpenGoal.resume?(request: request)
    prior = OpenGoal.current
    if prior && !prior[:request].to_s.strip.empty?
      request = prior[:request].to_s
      opts[:request] = request
      intent = request_intent(request: request)
    end
  elsif defined?(OpenGoal) && needs_host_work?(request: request) &&
        !%i[greeting howto recall].include?(intent)
    OpenGoal.begin!(request: request, session_id: session_id)
  end
  Thread.current[:pwn_request_intent] = intent
  Thread.current[:pwn_extinguished] = {}
  debug_progress(msg: "intent=#{intent} engine=#{engine}", debug: opts[:debug])
  expose_current_session(session_id: session_id)
  Mistakes.check_user_correction(request: request, session_id: session_id) if defined?(Mistakes)

  cheap = opts[:force_tools] != true && %i[greeting howto recall].include?(intent)

  if intent == :greeting && opts[:force_tools] != true
    debug_progress(msg: 'path=greeting', debug: opts[:debug])
    quiet_debug_tui!(debug: opts[:debug], reason: 'greeting')
    return answer_greeting(
      request: request,
      session_id: session_id
    )
  end

  # Thin system prompt only for remaining cheap paths (howto/recall).
  if cheap
    system_role_content = opts[:system_role_content]
    if system_role_content.nil? || system_role_content.to_s.empty?
      system_role_content = PWN::AI::Agent::PromptBuilder.build(
        session_id: session_id,
        request: request,
        thin: true
      )
      opts[:system_role_content] = system_role_content
    end
    if intent == :howto
      debug_progress(msg: 'path=howto', debug: opts[:debug])
      quiet_debug_tui!(debug: opts[:debug], reason: 'howto')
      return answer_howto(
        request: request,
        session_id: session_id,
        system_role_content: system_role_content
      )
    end
    if intent == :recall
      debug_progress(msg: 'path=recall', debug: opts[:debug])
      quiet_debug_tui!(debug: opts[:debug], reason: 'recall')
      return answer_recall(
        request: request,
        session_id: session_id,
        system_role_content: system_role_content
      )
    end
  end

  # --- act / recon / autonomous_goal: full context + tools ---
  # Reuse precomputed kind so TaskSummarizer.fresh does not classify twice.
  ts_state = (TaskSummarizer.fresh(request: request) if defined?(TaskSummarizer) && TaskSummarizer.enabled? && Thread.current[:pwn_reflect_depth].to_i.zero?)
  system_role_content = opts[:system_role_content] ||= PWN::AI::Agent::PromptBuilder.build(
    session_id: session_id,
    request: request
  )

  Registry.discover
  maybe_refresh_extro_snapshot!
  opts[:enabled_toolsets] = default_interactive_toolsets(request: request) unless opts.key?(:enabled_toolsets)

  # R5 — open the live MDP episode BEFORE the first Registry.rank so
  # Q(s,a) can advise this turn. Planning still owns the task list.
  if defined?(PWN::AI::Agent::Policy) && Policy.respond_to?(:begin_episode)
    Policy.begin_episode(
      session_id: session_id,
      request: request,
      intent: intent,
      engine: engine,
      ts_state: ts_state
    )
  end

  # Initial tool pool from the user request (bootstrap only). After
  # TaskSummarizer.emit_plan! we re-rank using English tangible tasks
  # so generated tasks — not the bare request — drive which tools
  # the model may call.
  # CORE_TOOLS is the default action space. Extra schemas are
  # opt-in via enabled_toolsets + core_only: false.
  core_only = opts.fetch(:core_only, true)
  tools = Registry.definitions(
    enabled: opts[:enabled_toolsets],
    relevance: request,
    core_only: core_only,
    intent: intent
  )
  messages = [{ role: 'system', content: system_role_content }]
  messages.concat(Learning.exemplars_for(request: request)) if local && defined?(Learning) && Learning.respond_to?(:exemplars_for)
  messages << { role: 'user', content: request }
  append_session(session_id: session_id, role: 'user', content: request)

  trivia = world_knowledge?(request: request)
  # Every host-work request gets a task compass. Trivia does not —
  # inventing English tasks there keeps the model from emitting a final.
  task_summary_plan!(state: ts_state, request: request, on_tool: on_tool) if defined?(TaskSummarizer) && !trivia
  # Re-bind tools from English plan so task list is the sole driver of
  # tool exposure/ranking (Registry keyword router + CORE).
  if ts_state.is_a?(Hash) && defined?(TaskSummarizer) && TaskSummarizer.respond_to?(:relevance_query)
    rq = TaskSummarizer.relevance_query(state: ts_state, request: request)
    unless rq.to_s.strip.empty?
      tools = Registry.definitions(
        enabled: opts[:enabled_toolsets],
        relevance: rq,
        core_only: core_only,
        intent: intent
      )
    end
  end
  # English-task-as-primary: inject tangible tasks only for host work.
  inject_task_focus!(messages: messages, state: ts_state, force: true, request: request) unless trivia
  predicted = nil
  Thread.current[:pwn_plan_predicted] = nil
  cal_state = calibration_state
  force_plan = cal_state[:force_plan]
  skip_plan = %i[howto recall greeting].include?(intent) || trivia
  if !skip_plan && (force_plan || agent_flag(key: :plan_first, default: local) || budget_exhaustion_hot?) && !Array(tools).empty?
    predicted = plan_first(messages: messages, request: request, ts_state: ts_state)
    # P22 — prefer explicit return; fall back to thread stash
    predicted = Thread.current[:pwn_plan_predicted] if predicted.nil?
    # unify_plan! may have rewritten English tasks — force refresh focus.
    # Re-rank tools from (possibly unified) English plan; never from
    # PLAN: tool-call scaffold jargon (unify_plan! refuses that).
    if ts_state.is_a?(Hash) && defined?(TaskSummarizer) && TaskSummarizer.respond_to?(:relevance_query)
      rq = TaskSummarizer.relevance_query(state: ts_state, request: request)
      unless rq.to_s.strip.empty?
        tools = Registry.definitions(
          enabled: opts[:enabled_toolsets],
          relevance: rq,
          core_only: core_only,
          intent: intent
        )
      end
    end
    inject_task_focus!(messages: messages, state: ts_state, force: true, request: request) unless trivia
  end
  if budget_exhaustion_hot?
    english_open = defined?(TaskSummarizer) && TaskSummarizer.respond_to?(:plan_open?) &&
                   TaskSummarizer.plan_open?(state: ts_state, messages: messages)
    hot_hint = if local_engine? && !english_open
                 '[pwn-ai/p17] Budget-exhaustion is the top open failure on this host. ' \
                   'Prefer the SHORTEST plan that finishes the ask (≤3 tool calls). ' \
                   'Emit a final answer as soon as you have evidence — do not explore.'
               else
                 '[pwn-ai/p17] Budget-exhaustion is the top open failure on this host. ' \
                   'Prefer the shortest plan that FULLY finishes the ask — no polite ' \
                   'handoffs, no exploration side-quests. Emit a final answer as soon ' \
                   'as you have evidence; keep going with tools until the goal is done ' \
                   'or truly blocked.'
               end
    messages << { role: 'user', content: hot_hint }
  end
  if force_plan && cal_state[:cal]
    messages << {
      role: 'user',
      content: "[pwn-ai/w3] engine=#{active_engine} is overconfident " \
               "(brier=#{cal_state[:cal][:brier]}, overconf=#{cal_state[:cal][:overconfidence]}). " \
               'Prefer high-judge exemplars, verify claims, and avoid speculative tool calls.'
    }
  end

  turn_fails = Hash.new(0)
  escalated  = false
  maybe_park_budget_scars!
  maybe_extinguish_parked!

  max_iters.times do |i|
    # 3.1 — compact history on local so tool dumps don't fill num_ctx
    compact_history!(messages: messages) if local
    # English-task-as-primary: when plan_idx advanced, tell the model
    # which plain-English task is active before the next tool batch.
    inject_task_focus!(messages: messages, state: ts_state, request: request) unless trivia

    # P17 — on the final iteration, strip tools and demand a plain-text
    # answer. Without this the model happily emits one more tool_calls
    # batch, burns the last slot, and lands on budget_exhausted with
    # nothing the user (or ORM) can use.
    # P17 deepen — when budget_hot, force text-only on the LAST TWO
    # iters so a final tool_calls batch cannot burn the terminal slot.
    # P17 deepen³ — under hot, force text-only on last THREE of the
    # 8-iter cap so a late tool binge cannot burn every salvage slot.
    # P17 structural: default hot text-only tail stays 3 (do NOT deepen to 4/6).
    # Plan-faithful headroom — short plan executing cleanly → delay strip to
    # last 1–2 so multi-step goals are not predestined to exhaust under cap 8.
    hot = budget_exhaustion_hot?
    plan_steps = begin
      predicted_plan = predicted || Thread.current[:pwn_plan_predicted]
      if predicted_plan.is_a?(Hash)
        Array(predicted_plan[:steps] || predicted_plan[:tools] || predicted_plan[:plan]).size
      elsif predicted_plan.is_a?(Array)
        predicted_plan.size
      else
        predicted_plan.to_s.scan(/\b(?:shell|pwn_eval|memory_|mistakes_|skill_|extro_|learning_|sessions_)\w*/).size
      end
    rescue StandardError
      0
    end
    # Plan-faithful: delay the text-only strip when a plan is executing
    # cleanly. Remote hot allows longer plans (runway 25); local hot
    # still favors short plans under the 8-iter cap.
    plan_step_limit = local_engine? ? 3 : 12
    plan_faithful = hot && plan_steps.positive? && plan_steps <= plan_step_limit &&
                    turn_fails['empty_final'].to_i.zero? &&
                    turn_fails.values.sum < 2
    # Last-iter strips tools only on the true last slot. English
    # leftovers and budget-hot must not steal runway from a live goal.
    text_only_iters = 1
    want_last = (i >= max_iters - text_only_iters)
    still_open = request_unsatisfied?(
      request: request,
      messages: messages,
      last_iter: false
    )
    last_iter = want_last && !still_open
    if last_iter
      tag = i >= max_iters - 1 ? 'FINAL ITERATION' : 'PENULTIMATE — wrap up'
      messages << {
        role: 'user',
        content: "[pwn-ai/p17] #{tag} — do NOT call any more tools. " \
                 'Write the complete answer from evidence already in this ' \
                 'transcript. If unfinished, say BLOCKED with evidence — not a ' \
                 'markdown outline or "# Remaining block". Do NOT ask the user ' \
                 'to confirm the next step.'
      }
    end

    msg = call_engine(messages: messages, tools: last_iter ? nil : tools, ts_state: ts_state)
    if msg.nil?
      task_summary_flush!(state: ts_state, on_tool: on_tool)
      quiet_debug_tui!(reason: 'engine_empty')
      return '[pwn-ai] engine returned no message'
    end

    calls = Array(msg[:tool_calls])
    text  = msg[:content].to_s

    # Belt-and-suspenders: plain-text shell(...) / tool forms from local
    # models under weak TEMPLATE {{ .Prompt }} become real tool_calls.
    if calls.empty? && !text.strip.empty? && !last_iter &&
       defined?(Dispatch) && Dispatch.respond_to?(:tool_calls_from_text)
      coerced = Dispatch.tool_calls_from_text(text: text)
      if coerced.any?
        wired = coerced.map { |tc| openai_wire_tool_call(tool_call: tc) }
        msg = msg.merge(tool_calls: wired, content: nil, _text_tool_coerced: true)
        calls = wired
        text = ''
        warn "[pwn-ai/loop] coerced #{wired.length} text tool call(s) on iter=#{i}" if local
      end
    end

    # Empty-final guard (local/thinking models): Ollama sometimes
    # returns done_reason=stop with eval_count<=1, empty content, no
    # tool_calls — historically surface as a blank TUI reply. Do NOT
    # commit that as the answer; drop the empty assistant turn,
    # inject a one-shot nudge, and keep iterating.
    if calls.empty? && text.strip.empty?
      warn "[pwn-ai/loop] empty final from #{engine} on iter=#{i}; nudging" if local
      messages << {
        role: 'user',
        content: 'Your previous reply was empty (no tool_calls and no content). ' \
                 'Either call a tool now, or write the final answer for the user as plain text. ' \
                 'Do not reply with an empty message.'
      }
      turn_fails['empty_final'] += 1
      next
    end

    messages << msg

    if calls.empty?
      # P28 — refuse polite mid-goal handoffs so multi-step tasks stay autonomous.
      if incomplete_final?(text: text, last_iter: false)
        turn_fails['incomplete_final'] += 1
        warn "[pwn-ai/loop] incomplete final on iter=#{i}; continuing autonomously"
        messages << {
          role: 'user',
          content: '[pwn-ai/p28] That reply was incomplete (handoff or narrated next step). ' \
                   'Do NOT monologue about what you will try. Do NOT ask the user to ' \
                   'confirm. Emit NATIVE tool_calls NOW (e.g. shell with a concrete ' \
                   'command). Never print shell(...) as plain text. Only emit a final ' \
                   'answer when the request is complete or truly blocked with evidence.'
        }
        next
      end
      unless may_finalize?(
        request: request,
        messages: messages,
        text: text
      )
        turn_fails['unsatisfied'] += 1
        warn "[pwn-ai/loop] original request not evidenced on iter=#{i}; continuing"
        messages << {
          role: 'user',
          content: '[pwn-ai] The original request is not evidenced yet. ' \
                   'Keep calling CORE_TOOLS (shell, pwn_eval) until that request is ' \
                   'done or truly blocked. Do not declare completion from a listing alone.'
        }
        next
      end
      debug_progress(msg: "final accepted chars=#{text.to_s.length}")
      quiet_debug_tui!(reason: 'final')
      append_session(session_id: session_id, role: 'assistant', content: text)
      Learning.auto_introspect(session_id: session_id, request: request, final: text, predicted: predicted, plan: ts_state && ts_state[:plan], ts_state: ts_state) if defined?(Learning) && should_auto_introspect?(local: local, turn_fails: turn_fails, iter: i)
      maybe_finish_policy(session_id: session_id, proxy_ok: true, ts_state: ts_state)
      task_summary_flush!(state: ts_state, on_tool: on_tool)
      OpenGoal.clear! if defined?(OpenGoal)
      return text
    end

    # One executive task brief for the whole collection, then the
    # individual tool lines. pwn-ai → task is one-to-many with tools.
    task_summary_about_to!(
      state: ts_state,
      tools: calls.map do |tool_call|
        {
          name: tool_call.dig(:function, :name).to_s,
          args: tool_call.dig(:function, :arguments)
        }
      end,
      request: request,
      on_tool: on_tool
    )

    calls.each do |tc|
      name    = tc.dig(:function, :name).to_s
      args    = tc.dig(:function, :arguments)
      entry   = Registry.lookup(name: name)
      started = Time.now
      if Thread.current[:pwn_extinguished].is_a?(Hash) && Thread.current[:pwn_extinguished][name]
        raw = JSON.generate(
          success: false,
          error: "extinguished_repeat: #{name} already failed this signature this turn — change args or tool",
          result: { stdout: '', stderr: "extinguished_repeat: #{name}", exit: 2 }
        )
      else
        raw = Dispatch.call(tool_call: tc)
      end
      debug_progress(msg: "dispatch #{name} ok_len=#{raw.to_s.length}")
      tele    = record_metrics(name: name, started: started, raw: raw, args: args, session_id: session_id, engine: engine, ts_state: ts_state)
      result  = Result.condition(content: raw, entry: entry)

      unless tele[:ok]
        fkey = Digest::SHA256.hexdigest("#{name}|#{args}")[0, 16]
        turn_fails[fkey] += 1
        persist = tele.dig(:mistake, :count).to_i
        count   = [turn_fails[fkey], persist].max
        hint    = defined?(Mistakes) ? Mistakes.correction_hint(tool: name, error: tele[:err] || raw[0, 300]) : ''
        # S2 — counterfactual A/B: at the repeat threshold, fork an
        # alt-persona branch, judge both, inject the winner. Real
        # advantage estimation; (loser, winner) → DPO preference.
        thresh = defined?(Mistakes) ? Mistakes::REPEAT_THRESHOLD : 3
        # P17 — never fork counterfactual when budget fingerprints dominate:
        # CF is another mini agent loop and is the #1 amplifier of
        # iteration-budget exhaustion on this host.
        if count >= thresh && !escalated && defined?(Curriculum) && !budget_exhaustion_hot?
          cf = (turn_fails["cf:#{fkey}"] += 1) == 1 ? Curriculum.counterfactual(request: request, name: name, args: args, error: tele[:err] || raw[0, 200], hint: hint) : nil
          hint = "#{hint}\n[pwn-ai/counterfactual] branch #{cf[:branch]} (score=#{cf[:score].round(2)}): #{cf[:content]}" if cf
        end
        result = guard_repeated_failure(name: name, count: count, hint: hint, result: result, mistake: tele[:mistake], args: args, shape: tele.dig(:mistake, :shape))
      end

      on_tool&.call(name, args, result)
      task_summary_record!(state: ts_state, name: name, args: args, result: result, on_tool: on_tool)

      messages << {
        role: 'tool',
        tool_call_id: tc[:id] || tc['id'] || "call_#{i}",
        name: name,
        content: result
      }
      append_session(
        session_id: session_id,
        role: 'tool',
        content: "#{name} → #{result[0, 1_024]}"
      )
    end

    # Do not inject "stop calling tools". Long goals keep CORE_TOOLS
    # until may_finalize? — the original request is the only signal.

    next unless local && !escalated && dispatch_fail_n(turn_fails: turn_fails) >= ESCALATE_AFTER_FAILS

    hint = escalate(request: request, turn_fails: turn_fails, session_id: session_id)
    if hint
      messages << { role: 'tool', tool_call_id: "escalation_#{i}", name: 'frontier_hint', content: hint }
      append_session(session_id: session_id, role: 'tool', content: "frontier_hint → #{hint[0, 1_024]}")
    end
    escalated = true
  end

  # P17 — exhaust path must still feed Learning so ORM/PRM/HER see the
  # failure (previously we only Mistakes.record'd and returned a bare
  # string — no session row, no judge, no hindsight).
  final_msg = '[pwn-ai] iteration budget exhausted'
  debug_progress(msg: 'final budget_exhausted')
  quiet_debug_tui!(reason: 'budget_exhausted')
  if defined?(Mistakes)
    Mistakes.record(
      tool: 'agent_loop',
      error: 'iteration budget exhausted without a final answer',
      session_id: session_id,
      source: :loop,
      shape: :budget_exhausted
    )
  end
  append_session(session_id: session_id, role: 'assistant', content: final_msg)
  Learning.auto_introspect(session_id: session_id, request: request, final: final_msg, predicted: predicted, plan: ts_state && ts_state[:plan], ts_state: ts_state) if defined?(Learning) && should_auto_introspect?(local: local, turn_fails: turn_fails, iter: max_iters)
  maybe_finish_policy(session_id: session_id, proxy_ok: false, ts_state: ts_state)
  task_summary_flush!(state: ts_state, on_tool: on_tool)
  final_msg
ensure
  TurnFinalizer.leave_user_path! if defined?(TurnFinalizer)
end

.world_knowledge?(opts = {}) ⇒ Boolean

True only when the ask needs a live host/file/browser effect. World-knowledge questions ("what color is a cherry") do not.

Returns:

  • (Boolean) 


381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
 
# File 'lib/pwn/ai/agent/loop.rb', line 381

public_class_method def self.world_knowledge?(opts = {})
  request = opts[:request].to_s.strip
  return false if request.empty?
  return false if request.length > 120
  return false if request.match?(ACT_REQUEST_RX)
  return false if request.match?(LOOKUP_REQUEST_RX)
  return false if request.match?(HOST_PATH_RX)
  return false if request.match?(BROWSER_REQUEST_RX)
  return false if request.match?(HOWTO_RX)
  return false if request.match?(%r{\b(this\s+(?:host|machine|box|system|subnet|file|repo)|/opt/|implement|scan|hosts?)\b}i)

  request.match?(/\A(?:what|why|who|when|where|which|how)\b/i)
rescue StandardError
  false
end