Agent API_p19

API (page 19 of 23)

Previous: API_p18.md

skills/hive_mind.py

Depends on: core/logging.py, modules/logging_config.py, skills/lazyown_groq_agents.py, skills/lazyown_llm.py, skills/lazyown_parquet_db.py Imported by: modules/loop_ai_bridge.py, modules/unified_dashboard.py, skills/lazyown_mcp.py, skills/swan_agent.py, skills/tests/test_hive_mind.py, skills/unified_orchestrator.py

• IReadableMemory.recall (method) skills/hive_mind.py:149 def recall(self, query, top_k) -- Return top_k items matching query.
• IWritableMemory.store (method) skills/hive_mind.py:157 def store(self, content) -- Persist content and return an opaque event_id.
• ICommandRunner.run (method) skills/hive_mind.py:175 def run(self, command, timeout) -- Execute command within timeout seconds and return text output.
• ICommandRunner.name (method) skills/hive_mind.py:180 def name(self) -- Human-readable identifier for this runner.
• EpisodicStore.__init__ (method) skills/hive_mind.py:215 def __init__(self, db_path)
• EpisodicStore.store (method) skills/hive_mind.py:235 def store(self, content, agent_id, role, event_type, meta, session_tag) -- Insert a row into hive_events and return event_id.
• EpisodicStore.recall (method) skills/hive_mind.py:274 def recall(self, query, top_k, role, event_type) -- Keyword search via FTS5.
• EpisodicStore.stats (method) skills/hive_mind.py:322 def stats(self) -- Return row counts for status reporting.
• EpisodicStore.forget (method) skills/hive_mind.py:329 def forget(self, older_than_hours, topic) -- Delete old or topic-matching rows.
• SemanticStore.__init__ (method) skills/hive_mind.py:358 def __init__(self, chroma_dir, episodic_fallback)
• SemanticStore.available (method) skills/hive_mind.py:380 def available(self) -- True when ChromaDB is operational.
• SemanticStore.store (method) skills/hive_mind.py:386 def store(self, content, agent_id, role, event_type, meta, session_tag, event_id) -- Add a document to ChromaDB.
• SemanticStore.recall (method) skills/hive_mind.py:434 def recall(self, query, top_k, where) -- Vector similarity search.
• SemanticStore.count (method) skills/hive_mind.py:477 def count(self) -- Return number of vectors stored.
• LongtermStore.store (method) skills/hive_mind.py:515 def store(self, content) -- Parquet store is read-only from here.
• LongtermStore.recall (method) skills/hive_mind.py:519 def recall(self, query, top_k) -- Keyword search across Parquet knowledge files.
• build_default_hive_memory (function) skills/hive_mind.py:558 def build_default_hive_memory(db_path) -- Factory that creates HiveMemory with the default three-layer backend.
• HiveMemory.__init__ (method) skills/hive_mind.py:582 def __init__(self, stores, episodic, semantic, longterm)
• HiveMemory.store (method) skills/hive_mind.py:596 def store(self, content, agent_id, role, event_type, meta, session_tag) -- Store content in all writable layers.
• HiveMemory.recall_episodic (method) skills/hive_mind.py:633 def recall_episodic(self, query, top_k, role, event_type) -- Keyword search across hive episodic memory.
• HiveMemory.recall_semantic (method) skills/hive_mind.py:647 def recall_semantic(self, query, top_k, where) -- Vector similarity search in ChromaDB.
• HiveMemory.recall_longterm (method) skills/hive_mind.py:660 def recall_longterm(self, query, top_k) -- Keyword search across Parquet knowledge files.
• HiveMemory.recall (method) skills/hive_mind.py:668 def recall(self, query, top_k) -- Unified recall: semantic (ChromaDB) + episodic (SQLite) + longterm (Parquet).
• HiveMemory.stats (method) skills/hive_mind.py:694 def stats(self) -- Return combined statistics from all layers.
• HiveMemory.forget (method) skills/hive_mind.py:704 def forget(self, older_than_hours, topic) -- Prune old or topic-matched entries.
• HiveBus.__init__ (method) skills/hive_mind.py:736 def __init__(self)
• HiveBus.publish (method) skills/hive_mind.py:740 def publish(self, msg) -- Deliver msg to recipient mailbox (and to "*" broadcast).
• HiveBus.receive (method) skills/hive_mind.py:746 def receive(self, agent_id, max_msgs) -- Drain mailbox for agent_id (includes broadcast).
• HiveBus.ack_broadcast (method) skills/hive_mind.py:754 def ack_broadcast(self, agent_id, msg_id) -- Mark a broadcast message as seen by agent_id (minimal bus — no-op).
• HiveBus.pending_count (method) skills/hive_mind.py:758 def pending_count(self, agent_id) -- Count messages waiting for agent_id.
• ConsensusProtocol.__init__ (method) skills/hive_mind.py:943 def __init__(self, risk_assessor)
• ConsensusProtocol.evaluate (method) skills/hive_mind.py:946 def evaluate(self, role, goal) -- Run the consensus vote for the given (role, goal) pair.
• DroneAgent.__init__ (method) skills/hive_mind.py:1091 def __init__(self, drone_id, role, goal, backend, memory, bus, max_iterations, api_key, model, runner, on_state_change)
• DroneAgent.start (method) skills/hive_mind.py:1128 def start(self) -- Spawn the drone thread.
• DroneAgent.join (method) skills/hive_mind.py:1137 def join(self, timeout) -- Block until the drone thread finishes.
• QueenBrain.__init__ (method) skills/hive_mind.py:1301 def __init__(self, memory, bus, pool, consensus)
• QueenBrain.plan (method) skills/hive_mind.py:1313 def plan(self, goal, n_drones) -- Decompose a high-level goal into per-drone task specs.
• QueenBrain.dispatch (method) skills/hive_mind.py:1332 def dispatch(self, tasks, backend, api_key, max_iterations) -- Spawn one drone per task in parallel.
• QueenBrain.plan_and_dispatch (method) skills/hive_mind.py:1405 def plan_and_dispatch(self, goal, n_drones, backend, api_key, max_iterations) -- Convenience: plan + dispatch in one call.
• QueenBrain.collect (method) skills/hive_mind.py:1417 def collect(self, drone_ids, timeout, poll_interval) -- Wait for all drones to finish (up to timeout seconds).
• QueenBrain.synthesize (method) skills/hive_mind.py:1447 def synthesize(self, drone_ids, original_goal) -- Read hive memory results for these drones and produce a synthesis summary.
• QueenBrain.read_bus (method) skills/hive_mind.py:1471 def read_bus(self) -- Drain queen's mailbox from drone result messages.
• DroneStateStore.__init__ (method) skills/hive_mind.py:1513 def __init__(self, db_path)
• DroneStateStore.upsert (method) skills/hive_mind.py:1523 def upsert(self, state) -- Insert or replace a drone state record.
• DroneStateStore.load_all (method) skills/hive_mind.py:1553 def load_all(self, limit) -- Load all persisted drone states ordered by most recently updated.
• DroneStateStore.mark_interrupted (method) skills/hive_mind.py:1588 def mark_interrupted(self) -- Called on daemon startup: any drone that was queued/running when the process died is now in an unknown state.
• DroneStateStore.load_interrupted (method) skills/hive_mind.py:1609 def load_interrupted(self) -- Return all drones currently in 'interrupted' state.
• DroneStateStore.delete_older_than (method) skills/hive_mind.py:1643 def delete_older_than(self, days) -- Prune records older than N days.
• DronePool.__init__ (method) skills/hive_mind.py:1673 def __init__(self, memory, bus, state_store)
• DronePool.recover_from_store (method) skills/hive_mind.py:1687 def recover_from_store(self) -- Load persisted drone states from a previous run.
• DronePool.spawn (method) skills/hive_mind.py:1708 def spawn(self, role, goal, backend, api_key, model, max_iterations, runner) -- Create and start a DroneAgent.
• DronePool.requeue_interrupted (method) skills/hive_mind.py:1741 def requeue_interrupted(self, backend, api_key, max_iterations) -- Re-spawn all drones previously marked 'interrupted'.
• DronePool.get_state (method) skills/hive_mind.py:1767 def get_state(self, drone_id) -- Return the current state snapshot for a drone, or None if unknown.
• DronePool.list_all (method) skills/hive_mind.py:1776 def list_all(self, limit) -- Return a sorted list of drone summary dicts (most recent first).
• DronePool.active_count (method) skills/hive_mind.py:1801 def active_count(self) -- Return number of live drones currently in queued or running state.
• HiveMind.__init__ (method) skills/hive_mind.py:1818 def __init__(self)
• HiveMind.spawn (method) skills/hive_mind.py:1829 def spawn(self, goal, role, backend, api_key, max_iterations) -- Spawn a single drone.
• HiveMind.spawn_hive (method) skills/hive_mind.py:1846 def spawn_hive(self, goal, n_drones, backend, api_key, max_iterations) -- Queen plans and dispatches multiple drones in parallel.
• HiveMind.status (method) skills/hive_mind.py:1863 def status(self) -- Return hive-wide status dict including historical drones from previous runs.
• HiveMind.recall (method) skills/hive_mind.py:1878 def recall(self, query, top_k) -- Unified recall from all memory layers.
• HiveMind.drone_result (method) skills/hive_mind.py:1882 def drone_result(self, drone_id) -- Return result dict for a specific drone.
• HiveMind.collect_and_synthesize (method) skills/hive_mind.py:1896 def collect_and_synthesize(self, drone_ids, goal) -- Synthesize results from multiple drones.
• HiveMind.forget (method) skills/hive_mind.py:1900 def forget(self, older_than_hours, topic) -- Prune hive memory.
• get_hive (function) skills/hive_mind.py:1911 def get_hive() -- Return the process-wide HiveMind singleton, creating it if necessary.
• mcp_hive_spawn (function) skills/hive_mind.py:1926 def mcp_hive_spawn(goal, role, n_drones, backend, max_iterations, api_key) -- Spawn one or more hive drones for a goal.
• mcp_hive_status (function) skills/hive_mind.py:1976 def mcp_hive_status() -- Return full hive status: active drones, memory stats, queen mailbox.
• mcp_hive_recall (function) skills/hive_mind.py:1981 def mcp_hive_recall(query, top_k) -- Semantic + episodic + longterm recall from hive memory.
• mcp_hive_plan (function) skills/hive_mind.py:1995 def mcp_hive_plan(goal, n_drones) -- Queen generates task decomposition without spawning drones.
• mcp_hive_result (function) skills/hive_mind.py:2004 def mcp_hive_result(drone_id) -- Get result from a specific drone.
• mcp_hive_collect (function) skills/hive_mind.py:2010 def mcp_hive_collect(drone_ids_csv, goal) -- Wait for drones to finish and return synthesized result. drone_ids_csv: comma-separated drone IDs.
• mcp_hive_forget (function) skills/hive_mind.py:2021 def mcp_hive_forget(older_than_hours, topic) -- Prune hive memory.
• mcp_hive_recover (function) skills/hive_mind.py:2027 def mcp_hive_recover(backend, api_key, max_iterations) -- Re-queue all drones that were interrupted by a previous process crash/restart.

skills/lazyown_automapper.py

Depends on: core/logging.py Imported by: skills/lazyown_mcp.py

• AutoMapper.__init__ (method) skills/lazyown_automapper.py:341 def __init__(self, lazyown_dir)
• AutoMapper.rescan (method) skills/lazyown_automapper.py:364 def rescan(self) -- Force a rescan of all source directories.
• AutoMapper.mcp_tools (method) skills/lazyown_automapper.py:368 def mcp_tools(self) -- Return list of mcp types.Tool objects for all discovered extensions.
• AutoMapper.dispatch (method) skills/lazyown_automapper.py:389 def dispatch(self, name, arguments, config, run_command_fn) -- Execute a dynamic tool.
• AutoMapper.update_skills_md (method) skills/lazyown_automapper.py:524 def update_skills_md(self, skills_md_path) -- Append / replace the auto-discovered section in skills/lazyown.md.
• AutoMapper.stats (method) skills/lazyown_automapper.py:571 def stats(self)
• AutoMapper.list_specs (method) skills/lazyown_automapper.py:581 def list_specs(self) -- Return raw spec dicts (useful for debugging / reporting).

skills/lazyown_campaign.py

Depends on: core/logging.py, modules/lesson_ingestor.py, modules/logging_config.py Imported by: skills/lazyown_mcp.py

• is_ip_in_scope (function) skills/lazyown_campaign.py:64 def is_ip_in_scope(ip, scope) -- Return True if ip belongs to any entry in scope.
• Campaign.is_active (method) skills/lazyown_campaign.py:148 def is_active(self) -- Return True when the campaign has not been completed yet.
• Campaign.to_dict (method) skills/lazyown_campaign.py:152 def to_dict(self) -- Serialise to a plain dict suitable for JSON persistence.
• Campaign.from_dict (method) skills/lazyown_campaign.py:157 def from_dict(cls, data) -- Deserialise from a dict loaded out of JSON.
• LessonLearned.to_dict (method) skills/lazyown_campaign.py:189 def to_dict(self)
• EpisodeReflectionEngine.__init__ (method) skills/lazyown_campaign.py:256 def __init__(self, hive_memory) -- Parameters hive_memory: Optional HiveMemory instance for semantic storage.
• EpisodeReflectionEngine.reflect (method) skills/lazyown_campaign.py:268 def reflect(self, campaign) -- Extract lessons from campaign and persist them.
• CampaignStore.__init__ (method) skills/lazyown_campaign.py:382 def __init__(self, campaign_file)
• CampaignStore.load (method) skills/lazyown_campaign.py:395 def load(self) -- Load the current campaign from disk.
• CampaignStore.create (method) skills/lazyown_campaign.py:414 def create(self, name, scope, notes) -- Create and persist a brand-new campaign.
• CampaignStore.complete (method) skills/lazyown_campaign.py:445 def complete(self, notes, run_reflection, hive_memory) -- Mark the active campaign as completed and optionally run reflection.
• CampaignStore.update_phase (method) skills/lazyown_campaign.py:500 def update_phase(self, host, phase) -- Set (or update) the attack phase for host.
• CampaignStore.add_milestone (method) skills/lazyown_campaign.py:516 def add_milestone(self, host, milestone_type, notes) -- Record a milestone achievement for host.
• CampaignStore.add_to_scope (method) skills/lazyown_campaign.py:540 def add_to_scope(self, ip_or_cidr) -- Append ip_or_cidr to the campaign scope (idempotent).
• CampaignStore.in_scope (method) skills/lazyown_campaign.py:559 def in_scope(self, ip) -- Return True if ip falls within the campaign scope.
• CampaignStore.summary (method) skills/lazyown_campaign.py:566 def summary(self) -- Return a human-readable multi-line summary of the campaign.
• main (function) skills/lazyown_campaign.py:667 def main(argv)

skills/lazyown_claudemd.py

Imported by: skills/lazyown_mcp.py, skills/tests/test_harness_e2e.py

• ClaudeMdLoader.__init__ (method) skills/lazyown_claudemd.py:32 def __init__(self, cwd)
• ClaudeMdLoader.load (method) skills/lazyown_claudemd.py:35 def load(self) -- Return merged instructions string from all applicable levels.
• ClaudeMdLoader.list_files (method) skills/lazyown_claudemd.py:90 def list_files(self) -- Return metadata for all CLAUDE.md files that exist.
• ClaudeMdLoader.create_user_file (method) skills/lazyown_claudemd.py:115 def create_user_file(self, content) -- Create/overwrite ~/.lazyown/CLAUDE.md.
• ClaudeMdLoader.create_project_file (method) skills/lazyown_claudemd.py:121 def create_project_file(self, content, local) -- Create CLAUDE.md (or CLAUDE.local.md) in cwd.
• ClaudeMdLoader.add_rule (method) skills/lazyown_claudemd.py:127 def add_rule(self, rule_name, content) -- Add a named rule file to .lazyown/rules/.
• ClaudeMdLoader.status_text (method) skills/lazyown_claudemd.py:135 def status_text(self)

skills/lazyown_context.py

Imported by: skills/lazyown_mcp.py, skills/lazyown_session.py, skills/tests/test_harness_e2e.py

• CompactionResult.summary (method) skills/lazyown_context.py:57 def summary(self)
• ContextCompactor.__init__ (method) skills/lazyown_context.py:72 def __init__(self, budget_override, snip_threshold, collapse_threshold)
• ContextCompactor.apply_budget (method) skills/lazyown_context.py:84 def apply_budget(self, content, tool_name) -- Hard-cap output.
• ContextCompactor.snip (method) skills/lazyown_context.py:99 def snip(self, content) -- Remove duplicate consecutive lines (log noise).
• ContextCompactor.microcompact (method) skills/lazyown_context.py:120 def microcompact(self, content) -- Strip known low-signal patterns.
• ContextCompactor.collapse (method) skills/lazyown_context.py:136 def collapse(self, content, tool_name) -- Replace repetitive sections with a structural summary.
• ContextCompactor.auto_compact_session (method) skills/lazyown_context.py:159 def auto_compact_session(entries) -- Template-based session summary when full pipeline is insufficient. entries: list of {"type": str, "data": dict} from...
• ContextCompactor.compact (method) skills/lazyown_context.py:203 def compact(self, content, tool_name) -- Apply all applicable layers in order.
• compact_output (function) skills/lazyown_context.py:237 def compact_output(content, tool_name) -- Convenience wrapper: compact and return string.

skills/lazyown_daemon.py

Depends on: core/logging.py, modules/event_engine.py, modules/logging_config.py, modules/session_state.py, modules/timeline_narrator.py, skills/sessions_watcher.py

• process_new_rows (function) skills/lazyown_daemon.py:108 def process_new_rows()
• file_watcher_loop (function) skills/lazyown_daemon.py:150 def file_watcher_loop(queue) -- Watch sessions/ for new/modified files and dispatch handlers.
• _Handler.on_created (method) skills/lazyown_daemon.py:201 def on_created(self, event)
• _Handler.on_modified (method) skills/lazyown_daemon.py:205 def on_modified(self, event)
• file_event_consumer (function) skills/lazyown_daemon.py:223 def file_event_consumer(queue) -- Consume FILE events from the queue and call _dispatch in executor.
• event_engine_loop (function) skills/lazyown_daemon.py:240 def event_engine_loop() -- Tail CSV, match rules, emit events — every ENGINE_INTERVAL_S seconds.
• heartbeat_loop (function) skills/lazyown_daemon.py:280 def heartbeat_loop() -- Emit HEARTBEAT event and write daemon_status.json every 30 seconds.
• toposwarm_keepalive_loop (function) skills/lazyown_daemon.py:332 def toposwarm_keepalive_loop() -- Keeps the TopoSwarm MCP server alive as a subprocess.
• cmd_run (function) skills/lazyown_daemon.py:453 def cmd_run() -- Run in foreground.
• cmd_start (function) skills/lazyown_daemon.py:462 def cmd_start() -- Fork, detach, and run in background.
• cmd_stop (function) skills/lazyown_daemon.py:495 def cmd_stop() -- Send SIGTERM to the running daemon.
• cmd_status (function) skills/lazyown_daemon.py:513 def cmd_status() -- Print status from daemon_status.json.

skills/lazyown_facts.py

Depends on: core/logging.py, modules/logging_config.py Imported by: cli/commands/mcp_bridge.py, lazyc2.py, skills/lazyown_groq_agents.py, skills/lazyown_llm.py, skills/lazyown_mcp.py, skills/sessions_watcher.py, skills/tests/test_facts.py

• Config.default (method) skills/lazyown_facts.py:45 def default(cls)
• HostFacts.highest_access (method) skills/lazyown_facts.py:145 def highest_access(self)
• HostFacts.open_ports (method) skills/lazyown_facts.py:156 def open_ports(self)
• HostFacts.services_by_name (method) skills/lazyown_facts.py:159 def services_by_name(self, name)
• INmapXmlParser.parse (method) skills/lazyown_facts.py:169 def parse(self, xml_path)
• ITextOutputParser.can_parse (method) skills/lazyown_facts.py:217 def can_parse(self, filename, content)
• ITextOutputParser.parse (method) skills/lazyown_facts.py:220 def parse(self, host, content, source_file)
• ITextOutputParser.parse_extended (method) skills/lazyown_facts.py:228 def parse_extended(self, host, content, source_file, port) -- Extended parse including vulnerabilities and web paths.
• CrackMapExecParser.can_parse (method) skills/lazyown_facts.py:250 def can_parse(self, filename, content)
• CrackMapExecParser.parse (method) skills/lazyown_facts.py:257 def parse(self, host, content, source_file)
• Enum4linuxParser.can_parse (method) skills/lazyown_facts.py:301 def can_parse(self, filename, content)
• Enum4linuxParser.parse (method) skills/lazyown_facts.py:304 def parse(self, host, content, source_file)
• SecretsdumpParser.can_parse (method) skills/lazyown_facts.py:341 def can_parse(self, filename, content)
• SecretsdumpParser.parse (method) skills/lazyown_facts.py:346 def parse(self, host, content, source_file)
• LdapParser.can_parse (method) skills/lazyown_facts.py:374 def can_parse(self, filename, content)
• LdapParser.parse (method) skills/lazyown_facts.py:377 def parse(self, host, content, source_file)
• KerbruteParser.can_parse (method) skills/lazyown_facts.py:403 def can_parse(self, filename, content)
• KerbruteParser.parse (method) skills/lazyown_facts.py:406 def parse(self, host, content, source_file)
• RpcclientParser.can_parse (method) skills/lazyown_facts.py:424 def can_parse(self, filename, content)
• RpcclientParser.parse (method) skills/lazyown_facts.py:427 def parse(self, host, content, source_file)
• GobusterFfufParser.can_parse (method) skills/lazyown_facts.py:447 def can_parse(self, filename, content)
• GobusterFfufParser.parse (method) skills/lazyown_facts.py:450 def parse(self, host, content, source_file)
• GobusterFfufParser.parse_extended (method) skills/lazyown_facts.py:453 def parse_extended(self, host, content, source_file, port)
• NiktoParser.can_parse (method) skills/lazyown_facts.py:502 def can_parse(self, filename, content)
• NiktoParser.parse (method) skills/lazyown_facts.py:505 def parse(self, host, content, source_file)
• NiktoParser.parse_extended (method) skills/lazyown_facts.py:508 def parse_extended(self, host, content, source_file, port)
• NucleiParser.can_parse (method) skills/lazyown_facts.py:545 def can_parse(self, filename, content)
• NucleiParser.parse (method) skills/lazyown_facts.py:548 def parse(self, host, content, source_file)
• NucleiParser.parse_extended (method) skills/lazyown_facts.py:551 def parse_extended(self, host, content, source_file, port)
• SslscanParser.can_parse (method) skills/lazyown_facts.py:581 def can_parse(self, filename, content)
• SslscanParser.parse (method) skills/lazyown_facts.py:584 def parse(self, host, content, source_file)
• SslscanParser.parse_extended (method) skills/lazyown_facts.py:587 def parse_extended(self, host, content, source_file, port)
• GenericOutputParser.can_parse (method) skills/lazyown_facts.py:635 def can_parse(self, filename, content)
• GenericOutputParser.parse (method) skills/lazyown_facts.py:638 def parse(self, host, content, source_file)
• FactStore.__init__ (method) skills/lazyown_facts.py:677 def __init__(self, cfg)
• FactStore.save (method) skills/lazyown_facts.py:727 def save(self)
• FactStore.ingest_xml (method) skills/lazyown_facts.py:791 def ingest_xml(self, xml_path) -- Parse one nmap XML file, merge results, return count of new facts.
• FactStore.ingest_text (method) skills/lazyown_facts.py:803 def ingest_text(self, txt_path, host_hint) -- Parse one tool output file, merge credential/share/access facts.
• FactStore.parse_all (method) skills/lazyown_facts.py:859 def parse_all(self, target) -- Scan sessions/ for all nmap XML and txt files.
• FactStore.get_host (method) skills/lazyown_facts.py:890 def get_host(self, host)
• FactStore.all_hosts (method) skills/lazyown_facts.py:893 def all_hosts(self)
• FactStore.context_for_command (method) skills/lazyown_facts.py:896 def context_for_command(self, host, category) -- Return a dict of substitution parameters for a command in the given attack category.
• FactStore.summary (method) skills/lazyown_facts.py:986 def summary(self, host) -- Return a human-readable summary of stored facts.
• create_tool_file (function) skills/lazyown_facts.py:1025 def create_tool_file(toolname, command, trigger, active, tools_dir) -- Write a new pwntomate .tool JSON file.
• main (function) skills/lazyown_facts.py:1074 def main()

skills/lazyown_groq_agents.py

Depends on: modules/atomic_enricher.py, modules/lazyown_bridge.py, modules/reactive_engine.py, modules/session_rag.py, modules/session_reader.py, modules/threat_model.py, skills/lazyown_facts.py, skills/lazyown_llm.py, skills/lazyown_mcp.py, skills/lazyown_objective.py, skills/lazyown_parquet_db.py Imported by: skills/hive_mind.py, skills/lazyown_mcp.py, skills/swan_agent.py, tests/test_bridge_catalog_filtered.py

• GroqAgentPool.__init__ (method) skills/lazyown_groq_agents.py:702 def __init__(self)
• GroqAgentPool.spawn (method) skills/lazyown_groq_agents.py:706 def spawn(self, goal, tools_filter, api_key, backend, max_iterations, system_prompt, block) -- Spawn a new agent.
• GroqAgentPool.status (method) skills/lazyown_groq_agents.py:778 def status(self, agent_id)
• GroqAgentPool.result (method) skills/lazyown_groq_agents.py:794 def result(self, agent_id)
• GroqAgentPool.list_all (method) skills/lazyown_groq_agents.py:807 def list_all(self, limit)
• get_pool (function) skills/lazyown_groq_agents.py:829 def get_pool()
• spawn_agent (function) skills/lazyown_groq_agents.py:838 def spawn_agent(goal, tools_filter, api_key, backend, max_iterations, block) -- Spawn a new Groq/Ollama agent.
• agent_status (function) skills/lazyown_groq_agents.py:857 def agent_status(agent_id)
• agent_result (function) skills/lazyown_groq_agents.py:861 def agent_result(agent_id)
• list_agents (function) skills/lazyown_groq_agents.py:865 def list_agents(limit)
• main (function) skills/lazyown_groq_agents.py:872 def main()

skills/lazyown_hooks.py

Depends on: cli/commands/enum.py Imported by: skills/lazyown_mcp.py, skills/tests/test_harness_e2e.py

• HookRegistry.__init__ (method) skills/lazyown_hooks.py:47 def __init__(self)
• HookRegistry.register (method) skills/lazyown_hooks.py:56 def register(self, event, handler)
• HookRegistry.run (method) skills/lazyown_hooks.py:59 def run(self, event, context) -- Run all handlers for event, chaining modified contexts.
• HookRegistry.metrics (method) skills/lazyown_hooks.py:79 def metrics(self) -- Return aggregate hook execution counts.
• sandbox_hook (function) skills/lazyown_hooks.py:101 def sandbox_hook(context) -- PRE_TOOL_USE: block catastrophic shell commands before permission evaluation.
• rate_limit_hook (function) skills/lazyown_hooks.py:116 def rate_limit_hook(context, window, limit) -- PRE_TOOL_USE: prevent rapid-fire C2 commands (>5 per 2s).
• audit_hook (function) skills/lazyown_hooks.py:139 def audit_hook(context, audit_path) -- AUDIT: append a one-line JSON record of every tool call to the audit log.
• timing_hook (function) skills/lazyown_hooks.py:161 def timing_hook(context) -- POST_TOOL_USE: annotate result with wall-clock execution time.
• start_timer_hook (function) skills/lazyown_hooks.py:176 def start_timer_hook(context) -- PRE_TOOL_USE: record start time for timing_hook.
• build_default_registry (function) skills/lazyown_hooks.py:187 def build_default_registry(sessions_dir) -- Create a HookRegistry with the built-in safety hooks wired up.
• get_registry (function) skills/lazyown_hooks.py:224 def get_registry(sessions_dir)

skills/lazyown_llm.py

Depends on: core/logging.py, modules/logging_config.py, skills/lazyown_facts.py, skills/lazyown_mcp.py, skills/lazyown_objective.py Imported by: skills/autonomous_daemon.py, skills/hive_mind.py, skills/lazyown_groq_agents.py, skills/lazyown_mcp.py

• LLMTool.openai_schema (method) skills/lazyown_llm.py:79 def openai_schema(self)
• LLMBridge.__init__ (method) skills/lazyown_llm.py:104 def __init__(self, backend, model, api_key)
• LLMBridge.register_tool (method) skills/lazyown_llm.py:118 def register_tool(self, name, description, parameters, func)
• LLMBridge.ask (method) skills/lazyown_llm.py:127 def ask(self, goal, context, max_iterations, system_prompt)
• run_command (function) skills/lazyown_llm.py:346 def run_command(command) -- Execute a LazyOwn shell command and return its output.
• read_nmap (function) skills/lazyown_llm.py:352 def read_nmap(target) -- Read the nmap scan file for the given target IP.
• read_plan (function) skills/lazyown_llm.py:369 def read_plan() -- Read the current attack plan from sessions/plan.txt.
• read_facts (function) skills/lazyown_llm.py:376 def read_facts(target) -- Return FactStore structured facts for a target (or all targets).
• read_objectives (function) skills/lazyown_llm.py:389 def read_objectives(limit) -- List the next N pending attack objectives.
• build_bridge (function) skills/lazyown_llm.py:438 def build_bridge(backend, model, api_key, with_default_tools) -- Create a ready-to-use LLMBridge with optional default LazyOwn tools.
• llm_ask (function) skills/lazyown_llm.py:468 def llm_ask(goal, context, backend, model, api_key, max_iterations, system_prompt, extra_tools) -- Single-call entry point for lazyown_mcp.py.
• main (function) skills/lazyown_llm.py:493 def main()

skills/lazyown_mcp.py

Depends on: cli/command_chain.py, cli/graph_advisor.py, cli/palette.py, cli/palette_command.py, cli/recommendation.py, cli/recommendation_signals.py, core/config.py, core/hardening.py, core/llm_budget.py, core/logging.py, core/payload_schema.py, modules/__init__.py, modules/ai_exploit_chain.py, modules/atomic_enricher.py, modules/atomic_exploit_bridge.py, modules/atomic_resolver.py, modules/auto_pivot.py, modules/auto_registry.py, modules/autonomous_exploit_engine.py, modules/backdoor/server.c, modules/chain_registry.py, modules/collab_bp.py, modules/cve_matcher.py, modules/dashboard_engine.py, modules/db.py, modules/evasion_engine.py, modules/event_bus.py, modules/event_engine.py, modules/evidence.py, modules/exploit_recommender.py, modules/exploitgym_gym.py, modules/goals.py, modules/integrations/misp_export.py, modules/intelligence_engine.py, modules/intercept.py, modules/killchain.py, modules/lazy_bridge.py, modules/lazy_engine.py, modules/lazy_loop.py, modules/lazy_mentions.py, modules/lazyown_bridge.py, modules/llm_client.py, modules/llm_evaluator.py, modules/llm_pool.py, modules/mcp_agent_bridge.py, modules/memory_store.py, modules/metrics.py, modules/module_registry.py, modules/obs_parser.py, modules/pipeline_engine.py, modules/planner.py, modules/playbook_engine.py, modules/reactive_engine.py, modules/recommender.py, modules/session_rag.py, modules/session_reader.py, modules/session_state.py, modules/threat_model.py, modules/timeline_narrator.py, modules/traffic_recorder.py, modules/ttpscope_bridge.py, modules/unified_dashboard.py, modules/world_model.py, scripts/lazy_opencode_config.py, skills/aci_planner.py, skills/autonomous_daemon.py, skills/autonomous_replay.py, skills/daemon_control.py, skills/hive_mind.py, skills/lazyown_automapper.py, skills/lazyown_campaign.py, skills/lazyown_claudemd.py, skills/lazyown_context.py, skills/lazyown_facts.py, skills/lazyown_groq_agents.py, skills/lazyown_hooks.py, skills/lazyown_llm.py, skills/lazyown_mcp_helpers.py, skills/lazyown_objective.py, skills/lazyown_parquet_db.py, skills/lazyown_permissions.py, skills/lazyown_policy.py, skills/lazyown_session.py, skills/mcp_lazy.py, skills/swan_agent.py Imported by: skills/autonomous_daemon.py, skills/lazyown_groq_agents.py, skills/lazyown_llm.py, skills/lazyown_mcp_opencode.py, skills/tests/test_mcp_smoke.py, tests/test_boyscout_contracts.py, tests/test_mcp_improvements.py, tests/test_rea_mcp.py

• register_handler (function) skills/lazyown_mcp.py:627 def register_handler(tool_name) -- Decorator that registers an async handler for tool_name.
• coerce_value (function) skills/lazyown_mcp.py:877 def coerce_value(k, v)
• list_tools (function) skills/lazyown_mcp.py:2124 def list_tools()
• substitute_playbook_target (function) skills/lazyown_mcp.py:6426 def substitute_playbook_target(command, target) -- Substitute the {target} token in a playbook step command.
• call_tool (function) skills/lazyown_mcp.py:6444 def call_tool(name, arguments)
• text (function) skills/lazyown_mcp.py:6458 def text(content)
• main (function) skills/lazyown_mcp.py:12848 def main()

skills/lazyown_mcp_helpers.py

Imported by: modules/c2_surfaces.py, skills/autonomous_daemon.py, skills/lazyown_mcp.py, tests/test_mcp_improvements.py

• is_likely_credential (function) skills/lazyown_mcp_helpers.py:37 def is_likely_credential(value) -- Score whether a string is plausibly a real credential.
• evidence_freshness (function) skills/lazyown_mcp_helpers.py:67 def evidence_freshness(path, threshold_seconds, now) -- Return age + stale flag for an evidence file.
• parse_task_value (function) skills/lazyown_mcp_helpers.py:125 def parse_task_value(title) -- Extract the embedded JSON-ish payload from a task title.
• audit_tasks (function) skills/lazyown_mcp_helpers.py:165 def audit_tasks(tasks, min_confidence) -- Classify each task as keep/drop with a reason.
• find_credential_provenance (function) skills/lazyown_mcp_helpers.py:196 def find_credential_provenance(value, sessions_dir, csv_name) -- Locate where a credential was first observed.
• evidence_grep (function) skills/lazyown_mcp_helpers.py:297 def evidence_grep(pattern, sessions_dir, scope, max_matches, max_file_bytes, case_insensitive) -- Grep through session artefacts.
• collect_pwntomate_evidence (function) skills/lazyown_mcp_helpers.py:376 def collect_pwntomate_evidence(rhost, sessions_dir) -- List pwntomate output dirs for a target with per-dir freshness.
• build_target_context (function) skills/lazyown_mcp_helpers.py:414 def build_target_context(host, port, sessions_dir, payload, world_model) -- Aggregate everything known about (host, port) into one structure.
• preflight_command (function) skills/lazyown_mcp_helpers.py:575 def preflight_command(command, payload, sessions_dir) -- Pre-flight a LazyOwn command without executing it.
• JobStore.__init__ (method) skills/lazyown_mcp_helpers.py:689 def __init__(self, max_jobs)
• JobStore.submit (method) skills/lazyown_mcp_helpers.py:694 def submit(self, command, runner, timeout) -- Start a runner(command, timeout) in a background thread.
• JobStore.status (method) skills/lazyown_mcp_helpers.py:727 def status(self, job_id)
• JobStore.list (method) skills/lazyown_mcp_helpers.py:732 def list(self, limit)
• take_snapshot (function) skills/lazyown_mcp_helpers.py:747 def take_snapshot(sessions_dir, payload, world_model, tasks) -- Capture a lightweight snapshot of the campaign state.
• diff_snapshot (function) skills/lazyown_mcp_helpers.py:792 def diff_snapshot(sessions_dir, payload, world_model, tasks) -- Return what changed since the last snapshot.
• needs_confirmation (function) skills/lazyown_mcp_helpers.py:875 def needs_confirmation(tool_name, arguments) -- Return True if tool should require an explicit confirm=True flag.

skills/lazyown_mcp_opencode.py

Depends on: modules/backdoor/server.c, skills/lazyown_mcp.py

• list_tools (function) skills/lazyown_mcp_opencode.py:62 def list_tools() -- Expose the curated subset under short names, keeping original schemas.
• call_tool (function) skills/lazyown_mcp_opencode.py:81 def call_tool(name, arguments) -- Proxy the short name to the full LazyOwn MCP handler.
• main (function) skills/lazyown_mcp_opencode.py:89 def main()

skills/lazyown_objective.py

Imported by: skills/autonomous_daemon.py, skills/lazyown_groq_agents.py, skills/lazyown_llm.py, skills/lazyown_mcp.py, skills/sessions_watcher.py, skills/tests/test_objectives.py, tests/test_daemon_loop.py

• Objective.sort_key (method) skills/lazyown_objective.py:116 def sort_key(self)
• ObjectiveStore.__init__ (method) skills/lazyown_objective.py:131 def __init__(self, path)
• ObjectiveStore.cleanup (method) skills/lazyown_objective.py:163 def cleanup(self) -- Expire pending objectives whose TTL has passed.
• ObjectiveStore.inject (method) skills/lazyown_objective.py:193 def inject(self, text, priority, source, context, notes)
• ObjectiveStore.complete (method) skills/lazyown_objective.py:243 def complete(self, obj_id, notes)
• ObjectiveStore.block (method) skills/lazyown_objective.py:246 def block(self, obj_id, reason)
• ObjectiveStore.skip (method) skills/lazyown_objective.py:249 def skip(self, obj_id, reason)
• ObjectiveStore.start (method) skills/lazyown_objective.py:252 def start(self, obj_id)
• ObjectiveStore.next_pending (method) skills/lazyown_objective.py:255 def next_pending(self)
• ObjectiveStore.list_pending (method) skills/lazyown_objective.py:261 def list_pending(self, limit)
• ObjectiveStore.list_all (method) skills/lazyown_objective.py:265 def list_all(self, status, limit)
• ObjectiveStore.summary (method) skills/lazyown_objective.py:271 def summary(self)
• decompose_goal (function) skills/lazyown_objective.py:296 def decompose_goal(goal_text, target) -- Split a compound goal into discrete verifiable objectives.
• read_soul (function) skills/lazyown_objective.py:320 def read_soul()
• write_soul (function) skills/lazyown_objective.py:326 def write_soul(content)
• current_plan (function) skills/lazyown_objective.py:330 def current_plan()
• full_context_for_claude (function) skills/lazyown_objective.py:336 def full_context_for_claude(target) -- Return a single dict with everything Claude Code needs to reason about the next objective: soul + pending...
• SoulUpdater.__init__ (method) skills/lazyown_objective.py:379 def __init__(self)
• SoulUpdater.update_phase (method) skills/lazyown_objective.py:417 def update_phase(self, phase) -- Update the Phase: line inside ## Current Focus.
• SoulUpdater.update_target (method) skills/lazyown_objective.py:421 def update_target(self, target) -- Update the Target: line inside ## Current Focus.
• SoulUpdater.update_os (method) skills/lazyown_objective.py:425 def update_os(self, os_name, target) -- Record detected operating system for a target.
• SoulUpdater.update_credentials (method) skills/lazyown_objective.py:432 def update_credentials(self, creds) -- Replace the ## Known Credentials section with discovered credentials.
• SoulUpdater.update_access (method) skills/lazyown_objective.py:459 def update_access(self, level, target, method) -- Update achieved access level for a target.
• SoulUpdater.update_vulnerabilities (method) skills/lazyown_objective.py:466 def update_vulnerabilities(self, vulns) -- Replace ## Key Vulnerabilities with the most severe findings.
• main (function) skills/lazyown_objective.py:485 def main()

skills/lazyown_parquet_db.py

Depends on: core/logging.py, modules/atomic_enricher.py, skills/lazyown_policy.py Imported by: cli/commands/mcp_bridge.py, modules/ttpscope_bridge.py, skills/hive_mind.py, skills/lazyown_groq_agents.py, skills/lazyown_mcp.py, skills/tests/test_parquet_db.py, skills/update_knowledge.py, tests/test_core_modules.py

• ParquetDB.__init__ (method) skills/lazyown_parquet_db.py:295 def __init__(self, lazyown_dir)
• ParquetDB.sync (method) skills/lazyown_parquet_db.py:316 def sync(self, csv_path) -- Ingest sessions/LazyOwn_session_report.csv into session_knowledge.parquet.
• ParquetDB.annotate (method) skills/lazyown_parquet_db.py:405 def annotate(self, row_id, success, category, outcome) -- Patch a row in session_knowledge.parquet by its id.
• ParquetDB.annotate_rich (method) skills/lazyown_parquet_db.py:434 def annotate_rich(self, row_id, output, finding_type, target_service, target_port, campaign_id, success, category... -- Full annotation after real execution: stores output snippet + all metadata.
• ParquetDB.query_session (method) skills/lazyown_parquet_db.py:490 def query_session(self, phase, target, success_only, limit) -- Query session_knowledge.parquet.
• ParquetDB.query_knowledge (method) skills/lazyown_parquet_db.py:526 def query_knowledge(self, keyword, parquet_name, columns, limit) -- Search for keyword across one or all parquets in parquets/.
• ParquetDB.query_atomic (method) skills/lazyown_parquet_db.py:576 def query_atomic(self, keyword, mitre_id, platform, scope, has_prereqs, complexity, limit, include_command) -- Structured query over the enriched Atomic Red Team catalogue (parquets/techniques_enriched.parquet).
• ParquetDB.context_for_phase (method) skills/lazyown_parquet_db.py:630 def context_for_phase(self, phase, target, limit) -- Return a rich context dict for MCP reasoning:
• ParquetDB.stats (method) skills/lazyown_parquet_db.py:729 def stats(self)
• ParquetDB.list_parquets (method) skills/lazyown_parquet_db.py:743 def list_parquets(self)
• ParquetDB.train_classifier (method) skills/lazyown_parquet_db.py:748 def train_classifier(self, min_rows) -- Train a lightweight RandomForest classifier on session_knowledge.parquet.
• ParquetDB.predict_success (method) skills/lazyown_parquet_db.py:843 def predict_success(self, command, category) -- Use the trained classifier to estimate probability of success.
• verify_model_integrity (function) skills/lazyown_parquet_db.py:892 def verify_model_integrity(model_path, encoder_path, hash_path) -- Verify SHA256 integrity of classifier model and encoder files.
• get_pdb (function) skills/lazyown_parquet_db.py:948 def get_pdb(lazyown_dir)
• main (function) skills/lazyown_parquet_db.py:961 def main()

skills/lazyown_permissions.py

Depends on: cli/commands/enum.py Imported by: skills/lazyown_mcp.py, skills/tests/test_harness_e2e.py

• PermissionRule.to_dict (method) skills/lazyown_permissions.py:36 def to_dict(self)
• PermissionSystem.__init__ (method) skills/lazyown_permissions.py:103 def __init__(self, sessions_dir)
• PermissionSystem.add_rule (method) skills/lazyown_permissions.py:145 def add_rule(self, tool_pattern, action, condition, description)
• PermissionSystem.remove_rule (method) skills/lazyown_permissions.py:151 def remove_rule(self, tool_pattern, action)
• PermissionSystem.set_mode (method) skills/lazyown_permissions.py:157 def set_mode(self, mode)
• PermissionSystem.list_rules (method) skills/lazyown_permissions.py:162 def list_rules(self)
• PermissionSystem.evaluate (method) skills/lazyown_permissions.py:167 def evaluate(self, tool_name, arguments) -- Returns (decision, reason). decision: "allow" | "deny" | "ask"
• PermissionSystem.metrics (method) skills/lazyown_permissions.py:237 def metrics(self) -- Aggregate audit log into per-decision and per-tool counts.
• PermissionSystem.status_text (method) skills/lazyown_permissions.py:279 def status_text(self)

skills/lazyown_policy.py

Depends on: cli/commands/enum.py, core/logging.py, modules/detection_oracle.py, modules/logging_config.py Imported by: cli/recommendation_signals.py, modules/unified_dashboard.py, skills/autonomous_daemon.py, skills/lazyown_mcp.py, skills/lazyown_parquet_db.py, skills/swan_agent.py, tests/test_engage_orchestrator.py, tests/test_moe_rl_swan.py, tests/test_scope_bound_auto_gate.py

• Config.default (method) skills/lazyown_policy.py:82 def default(cls) -- Construct configuration anchored to this file's repository root.
• Config.ollama_url (method) skills/lazyown_policy.py:123 def ollama_url(self) -- Full URL for the Ollama generate endpoint.
• infer_category (function) skills/lazyown_policy.py:267 def infer_category(command, args) -- Return the ActionCategory best matching the command and its arguments.
• ClassificationResult.unknown (method) skills/lazyown_policy.py:317 def unknown(cls, tier) -- Produce a result indicating the tier could not classify.
• EpisodeRecord.to_dict (method) skills/lazyown_policy.py:357 def to_dict(self) -- Serialise to a plain dict suitable for JSON encoding.
• EpisodeRecord.from_dict (method) skills/lazyown_policy.py:362 def from_dict(cls, d) -- Deserialise from a dict produced by to_dict.
• IOutputClassifier.classify (method) skills/lazyown_policy.py:376 def classify(self, command, args, output, exit_code) -- Classify the outcome of one command execution.
• ExitCodeClassifier.classify (method) skills/lazyown_policy.py:401 def classify(self, command, args, output, exit_code) -- Return a low-confidence result based solely on the exit code.
• HeuristicClassifier.classify (method) skills/lazyown_policy.py:444 def classify(self, command, args, output, exit_code) -- Scan output for known success or failure signatures.
• _OllamaClassifierBase.__init__ (method) skills/lazyown_policy.py:496 def __init__(self, cfg, model, tier_name)
• _OllamaClassifierBase.classify (method) skills/lazyown_policy.py:570 def classify(self, command, args, output, exit_code) -- Send the command context to Ollama and parse the JSON reply.
• OllamaSmallClassifier.__init__ (method) skills/lazyown_policy.py:591 def __init__(self, cfg)
• OllamaLargeClassifier.__init__ (method) skills/lazyown_policy.py:601 def __init__(self, cfg)
• UserInteractiveClassifier.classify (method) skills/lazyown_policy.py:616 def classify(self, command, args, output, exit_code) -- Prompt the operator and return their authoritative verdict.
• CascadeClassifier.__init__ (method) skills/lazyown_policy.py:666 def __init__(self, cfg, interactive)
• CascadeClassifier.classify (method) skills/lazyown_policy.py:677 def classify(self, command, args, output, exit_code) -- Return the first sufficiently confident result, or the best available.
• DetectionRiskAssessor.__init__ (method) skills/lazyown_policy.py:715 def __init__(self)
• DetectionRiskAssessor.assess_probability (method) skills/lazyown_policy.py:733 def assess_probability(self, command, args, category) -- Return detection probability in [0.0, 1.0].
• DetectionRiskAssessor.is_high_risk (method) skills/lazyown_policy.py:746 def is_high_risk(self, command, args, category) -- Return True when detection probability meets or exceeds the threshold.
• RewardCalculator.__init__ (method) skills/lazyown_policy.py:773 def __init__(self, cfg, risk_assessor)
• RewardCalculator.calculate (method) skills/lazyown_policy.py:783 def calculate(self, category, outcome) -- Return the reward integer for the given (category, outcome) pair.
• RewardCalculator.calculate_with_detection (method) skills/lazyown_policy.py:798 def calculate_with_detection(self, category, outcome, command, args) -- Return (reward, detection_probability) for the given execution.
• CSVSessionReader.__init__ (method) skills/lazyown_policy.py:850 def __init__(self, cfg)
• CSVSessionReader.read (method) skills/lazyown_policy.py:855 def read(self) -- Parse the CSV and return all classifiable rows as StepRecord instances.
• IEpisodeStore.append_step (method) skills/lazyown_policy.py:903 def append_step(self, step) -- Add a step to the episode for its target, creating the episode if needed.
• IEpisodeStore.load_all (method) skills/lazyown_policy.py:907 def load_all(self) -- Load every stored episode.
• IEpisodeStore.get_episode (method) skills/lazyown_policy.py:911 def get_episode(self, target) -- Return the episode for the given target, or None.
• JSONLEpisodeStore.__init__ (method) skills/lazyown_policy.py:923 def __init__(self, cfg)
• JSONLEpisodeStore.append_step (method) skills/lazyown_policy.py:927 def append_step(self, step) -- Upsert the step into its target's episode and persist.
• JSONLEpisodeStore.load_all (method) skills/lazyown_policy.py:947 def load_all(self) -- Deserialise all episodes from the .jsonl file.
• JSONLEpisodeStore.get_episode (method) skills/lazyown_policy.py:963 def get_episode(self, target) -- Return the episode for target, or None if not found.
• TransitionTable.__init__ (method) skills/lazyown_policy.py:987 def __init__(self, cfg)
• TransitionTable.record (method) skills/lazyown_policy.py:993 def record(self, from_state, to_category, outcome) -- Increment the counter for the given transition and persist.
• TransitionTable.query (method) skills/lazyown_policy.py:1003 def query(self, from_state) -- Return (next_category, success_rate, total_count) tuples for the given from_state, filtered by min_transition_count...
• PolicyEngine.__init__ (method) skills/lazyown_policy.py:1107 def __init__(self, cfg, transitions)
• PolicyEngine.recommend (method) skills/lazyown_policy.py:1111 def recommend(self, recent_steps) -- Return up to top_k dicts with keys: category, reason, confidence, source.
• SessionClassificationPipeline.__init__ (method) skills/lazyown_policy.py:1202 def __init__(self, cfg, interactive)
• SessionClassificationPipeline.process (method) skills/lazyown_policy.py:1210 def process(self, target, command, args, output, exit_code, timestamp) -- Classify, reward, store, and return a StepRecord for the given execution.
• PolicyAdvisor.__init__ (method) skills/lazyown_policy.py:1267 def __init__(self, cfg)
• PolicyAdvisor.advise (method) skills/lazyown_policy.py:1272 def advise(self, target) -- Return the policy engine's recommendations for the given target.
• PolicyAdvisor.episode_summary (method) skills/lazyown_policy.py:1277 def episode_summary(self, target) -- Return a high-level summary dict for the target's episode.
• LazyOwnPolicyIntegration.__init__ (method) skills/lazyown_policy.py:1302 def __init__(self, cfg)
• LazyOwnPolicyIntegration.on_command_complete (method) skills/lazyown_policy.py:1307 def on_command_complete(self, target, command, args, output, exit_code) -- Classify a completed command and record it in the episode store.
• LazyOwnPolicyIntegration.get_recommendations (method) skills/lazyown_policy.py:1318 def get_recommendations(self, target) -- Return ranked next-action recommendations for the target.
• ApprovalOutcome.is_approved (method) skills/lazyown_policy.py:1365 def is_approved(self) -- True when the gate authorised execution.
• ApprovalOutcome.is_denied (method) skills/lazyown_policy.py:1370 def is_denied(self) -- True when the gate denied execution.
• IApprovalSink.announce (method) skills/lazyown_policy.py:1384 def announce(self, request) -- Publish the request so operators can see and resolve it.
• IApprovalSink.resolution_for (method) skills/lazyown_policy.py:1388 def resolution_for(self, approval_id) -- Return the latest resolution for the request or None if pending.
• FileApprovalSink.__init__ (method) skills/lazyown_policy.py:1400 def __init__(self, sessions_dir)
• FileApprovalSink.path (method) skills/lazyown_policy.py:1405 def path(self) -- Absolute path of the underlying JSONL file.
• FileApprovalSink.announce (method) skills/lazyown_policy.py:1409 def announce(self, request)
• FileApprovalSink.resolution_for (method) skills/lazyown_policy.py:1426 def resolution_for(self, approval_id)
• BroadcastApprovalSink.__init__ (method) skills/lazyown_policy.py:1463 def __init__(self, narrator, file_sink)
• BroadcastApprovalSink.announce (method) skills/lazyown_policy.py:1471 def announce(self, request)
• BroadcastApprovalSink.resolution_for (method) skills/lazyown_policy.py:1499 def resolution_for(self, approval_id)
• StdinApprovalSink.__init__ (method) skills/lazyown_policy.py:1513 def __init__(self, stream)
• StdinApprovalSink.announce (method) skills/lazyown_policy.py:1523 def announce(self, request)
• StdinApprovalSink.resolution_for (method) skills/lazyown_policy.py:1552 def resolution_for(self, approval_id)
• CompositeApprovalSink.__init__ (method) skills/lazyown_policy.py:1564 def __init__(self, sinks)
• CompositeApprovalSink.announce (method) skills/lazyown_policy.py:1569 def announce(self, request)
• CompositeApprovalSink.resolution_for (method) skills/lazyown_policy.py:1576 def resolution_for(self, approval_id)
• ApprovalGate.__init__ (method) skills/lazyown_policy.py:1634 def __init__(self, sink, payload_path, poll_interval_s, poll_timeout_s, sleep_fn)
• ApprovalGate.request (method) skills/lazyown_policy.py:1667 def request(self, target, phase, command, reason) -- Return an ApprovalOutcome for the proposed execution.
• ScopeBoundAutoGate.__init__ (method) skills/lazyown_policy.py:1850 def __init__(self, payload_path, in_scope_fn)
• ScopeBoundAutoGate.request (method) skills/lazyown_policy.py:1866 def request(self, target, phase, command, reason) -- Return an ApprovalOutcome without ever prompting a human.
• HistoryBootstrapper.__init__ (method) skills/lazyown_policy.py:1928 def __init__(self, cfg)
• HistoryBootstrapper.run (method) skills/lazyown_policy.py:1934 def run(self) -- Process all CSV rows and return the total number of steps indexed.
• main (function) skills/lazyown_policy.py:2036 def main() -- Parse CLI arguments and dispatch to the appropriate command handler.

skills/lazyown_session.py

Depends on: skills/lazyown_context.py Imported by: skills/lazyown_mcp.py, skills/tests/test_harness_e2e.py

• SessionTranscript.__init__ (method) skills/lazyown_session.py:82 def __init__(self, sessions_dir, session_id)
• SessionTranscript.append (method) skills/lazyown_session.py:106 def append(self, event_type, data) -- Append a single event.
• SessionTranscript.get_recent (method) skills/lazyown_session.py:143 def get_recent(self, n) -- Return the last n events.
• SessionTranscript.get_all (method) skills/lazyown_session.py:161 def get_all(self)
• SessionTranscript.count (method) skills/lazyown_session.py:164 def count(self)
• SessionTranscript.maybe_auto_compact (method) skills/lazyown_session.py:174 def maybe_auto_compact(self, threshold) -- If event count since the last compact_boundary exceeds threshold, generate a template summary of those events and...
• SessionTranscript.add_compact_boundary (method) skills/lazyown_session.py:209 def add_compact_boundary(self, summary, preserved_uuids) -- Mark a compaction point.
• SessionTranscript.fork (method) skills/lazyown_session.py:227 def fork(self, new_id) -- Create a new transcript forked from this one.
• SessionTranscript.status_text (method) skills/lazyown_session.py:257 def status_text(self)
• get_transcript (function) skills/lazyown_session.py:281 def get_transcript(sessions_dir, session_id) -- Return the active transcript singleton.
• reset_transcript (function) skills/lazyown_session.py:294 def reset_transcript(sessions_dir, session_id) -- Force a new transcript (e.g., after fork or session reset).

skills/mcp_generated_tools.py

• get_generated_tool_definitions (function) skills/mcp_generated_tools.py:659 def get_generated_tool_definitions() -- Return list[types.Tool] entries for all generated tools.
• register_all_generated_handlers (function) skills/mcp_generated_tools.py:684 def register_all_generated_handlers(register_handler_fn, make_text_fn, run_lazyown_cmd_fn) -- Register every generated handler via register_handler_fn.

skills/mcp_lazy.py

Imported by: skills/lazyown_mcp.py, tests/test_mcp_lazy.py

• ensure_optional (function) skills/mcp_lazy.py:18 def ensure_optional(namespace, flag, loader) -- Return the cached flag or run loader once to compute it.

skills/mcp_tool_generator.py

• load_command_index (function) skills/mcp_tool_generator.py:29 def load_command_index()
• extract_commands (function) skills/mcp_tool_generator.py:34 def extract_commands(index) -- Return {command_name: summary} for all unique non-duplicate commands.
• extract_existing_handlers (function) skills/mcp_tool_generator.py:55 def extract_existing_handlers() -- Parse lazyown_mcp.py for all defined MCP tool names.
• sanitize_description (function) skills/mcp_tool_generator.py:72 def sanitize_description(summary, max_len) -- Truncate and clean a command summary for use as a tool description.
• sanitize_docstring (function) skills/mcp_tool_generator.py:85 def sanitize_docstring(text) -- Escape backslashes and quotes for triple-quoted docstrings.
• generate_module (function) skills/mcp_tool_generator.py:106 def generate_module(commands, existing_handlers, limit) -- Build the complete Python source for mcp_generated_tools.py.
• verify_coverage (function) skills/mcp_tool_generator.py:203 def verify_coverage(commands, existing_handlers) -- Compare indexed commands against existing MCP tool coverage.
• main (function) skills/mcp_tool_generator.py:220 def main()

skills/sessions_watcher.py

Depends on: core/logging.py, modules/event_engine.py, modules/logging_config.py, skills/lazyown_facts.py, skills/lazyown_objective.py Imported by: skills/lazyown_daemon.py

• _Handler.on_created (method) skills/sessions_watcher.py:376 def on_created(self, event)
• _Handler.on_modified (method) skills/sessions_watcher.py:380 def on_modified(self, event)
• main (function) skills/sessions_watcher.py:430 def main()

skills/setup.sh

• info (function) skills/setup.sh:15
• warn (function) skills/setup.sh:16
• die (function) skills/setup.sh:17
• paths_for_sudo (function) skills/setup.sh:61 -- Collect paths that exist on this system
• add_nopasswd (function) skills/setup.sh:92

skills/swan_agent.py

Depends on: core/logging.py, modules/detection_oracle.py, modules/logging_config.py, modules/moe_router.py, modules/rl_trainer.py, skills/hive_mind.py, skills/lazyown_groq_agents.py, skills/lazyown_policy.py Imported by: modules/loop_ai_bridge.py, skills/autonomous_daemon.py, skills/lazyown_mcp.py, skills/unified_orchestrator.py, tests/test_moe_rl_swan.py

• SwanResult.is_success (method) skills/swan_agent.py:167 def is_success(self)
• IResultAggregator.aggregate (method) skills/swan_agent.py:209 def aggregate(self, task_type, goal, votes) -- Return (synthesis_text, consensus_confidence).
• ISwanOrchestrator.run (method) skills/swan_agent.py:226 def run(self, task_type, goal, engagement_phase, timeout) -- Execute with the best single expert and return result.
• ISwanOrchestrator.ensemble_run (method) skills/swan_agent.py:236 def ensemble_run(self, task_type, goal, n_experts, engagement_phase, timeout) -- Execute with top-N experts in parallel and synthesize results.
• WeightedTextAggregator.aggregate (method) skills/swan_agent.py:264 def aggregate(self, task_type, goal, votes)
• ExpertExecutor.execute (method) skills/swan_agent.py:351 def execute(self, expert_id, backend, model, goal, task_type, timeout, api_key) -- Run the expert and return a SwanResult.
• OutcomeEvaluator.evaluate (method) skills/swan_agent.py:479 def evaluate(self, result, task_type) -- Return (reward, detection_prob).
• SwanOrchestrator.__init__ (method) skills/swan_agent.py:572 def __init__(self, executor, evaluator, aggregator, api_key)
• SwanOrchestrator.run (method) skills/swan_agent.py:592 def run(self, task_type, goal, engagement_phase, timeout) -- Execute the task with the best single expert selected by the MoE+RL router.
• SwanOrchestrator.ensemble_run (method) skills/swan_agent.py:680 def ensemble_run(self, task_type, goal, n_experts, engagement_phase, timeout) -- Execute with top-N experts in parallel and synthesize results.
• SwanOrchestrator.status (method) skills/swan_agent.py:763 def status(self) -- Return a diagnostic snapshot of the SWAN system.
• mcp_swan_run (function) skills/swan_agent.py:945 def mcp_swan_run(task_type, goal, phase) -- Route and execute a task with the best MoE+RL-selected expert.
• mcp_swan_ensemble (function) skills/swan_agent.py:968 def mcp_swan_ensemble(task_type, goal, n_experts, phase) -- Run top-N experts in parallel and return a synthesized result.
• mcp_swan_status (function) skills/swan_agent.py:996 def mcp_swan_status() -- Return SWAN system status: expert weights, RL epsilon, performance.
• mcp_swan_route (function) skills/swan_agent.py:1003 def mcp_swan_route(task_type, goal) -- Show which expert would be selected for a task without executing.
• get_swan (function) skills/swan_agent.py:1039 def get_swan(api_key) -- Return (or create) the module-level singleton SwanOrchestrator.

skills/toposwarm_autonomous.py

Depends on: core/logging.py, modules/logging_config.py, modules/reactive_engine.py, modules/toposwarm_bridge.py

• PentestState.add_finding (method) skills/toposwarm_autonomous.py:145 def add_finding(self, phase_name, text)
• PentestState.summary (method) skills/toposwarm_autonomous.py:149 def summary(self)
• PentestState.to_dict (method) skills/toposwarm_autonomous.py:160 def to_dict(self)
• AutonomousAgent.__init__ (method) skills/toposwarm_autonomous.py:310 def __init__(self, state, no_model, verbose, effort, max_phases, json_out)
• AutonomousAgent.run (method) skills/toposwarm_autonomous.py:432 def run(self, start_phase)
• main (function) skills/toposwarm_autonomous.py:489 def main(argv)

Next: API_p20.md