Repository navigation
Agent INDEX_p2
grisun0 edited this page Oct 16, 2026
·
21 revisions
Previous: INDEX.md
| File | Purpose | Subsystem | Symbols | Used by |
|---|---|---|---|---|
modules/lazyreverse_shell.sh |
Nombre del script: lazyreverse_shell.sh Autor: Gris Iscomeback Correo electrónico... | modules | 2 | 0 |
modules/lazyrtpflood.sh |
- | modules | 0 | 0 |
modules/lazyvpnshield.sh |
Author: Nisrin Ahmed aka Wh1teDrvg0n Modify: grisun0 | modules | 7 | 0 |
modules/lazywps.sh |
- | modules | 1 | 0 |
modules/legacy/__init__.py |
Backward-compatibility shims for deprecated scripts. | misc | 0 | 0 |
modules/lesson_ingestor.py |
Lesson ingestion bridge between EpisodeReflectionEngine and MoERouter. | modules | 11 | 2 |
modules/lilsplunky.py |
Author: Your Name Email: youremail@example.com Creation Date: 14/04/2025 License: GPL v3... | modules | 14 | 0 |
modules/linux_advanced_payloads.py |
Advanced Linux payloads — LD_PRELOAD rootkits, eBPF, PAM backdoors, kernel implants. | modules | 17 | 1 |
modules/listener_manager.py |
Multi-listener manager for LazyOwn C2. | modules | 22 | 2 |
modules/live_surface.py |
Live attack-surface graph derived from the world model. | modules | 7 | 3 |
modules/llm_adapter.py |
LLM adapter facade — single import point for all LLM backends used by lazyc2. | modules | 15 | 6 |
modules/llm_client.py |
LazyOwn Unified LLM Client | modules | 12 | 5 |
modules/llm_evaluator.py |
— Records LLM decisions and their outcomes, computes quality metrics, and exports fine-tuning... | modules | 28 | 2 |
modules/llm_factory.py |
LLM backend factory and selection utilities. | modules | 28 | 27 |
modules/llm_pool.py |
Ordered LLM failover pool with health-aware skipping. | modules | 20 | 8 |
modules/llm_prompts.py |
Canonical prompt-template and knowledge-base contract for LLM consumers. | modules | 28 | 4 |
modules/log_tamper.py |
Log tampering — Windows Event Log, Linux journald/auditd, macOS unified log. | modules | 9 | 1 |
modules/logging_config.py |
Centralized logging configuration for the LazyOwn framework. | modules | 21 | 24 |
modules/loop_ai_bridge.py |
Single edge between the autonomous loop and the LazyOwn AI stack. | modules | 23 | 5 |
modules/loop_progress.py |
Rich Live milestone panel for the autonomous loop. | modules | 15 | 4 |
modules/macos_payloads.py |
macOS payload generation — .app bundles, persistence, TCC bypass, Swift/ObjC. | modules | 16 | 1 |
modules/mario.py |
- | modules | 1 | 0 |
modules/mcp_agent_bridge.py |
LazyOwn MCP Agent Bridge | modules | 21 | 2 |
modules/memory_cleaner.py |
Memory artifact cleanup — process memory wipe, environment variable scrub, clipboard clear. | modules | 7 | 1 |
modules/memory_store.py |
— Episodic memory for the LazyOwn auto_loop. | modules | 33 | 2 |
modules/metrics.py |
LazyOwn metrics surface. | modules | 24 | 6 |
modules/mfa_bypass.py |
MFA Bypass Toolkit — techniques for circumventing multi-factor authentication. | modules | 10 | 1 |
modules/mkcloudflaretunnel.sh |
Download cloudflared and create a tunnel to localhost on a specified port. | modules | 0 | 0 |
modules/module_registry.py |
Unified module registry for LazyOwn — catalog, search, use/run workflow. | modules | 29 | 6 |
modules/moe_router.py |
modules/moe_router.py | modules | 39 | 6 |
modules/morse.py |
Morse code conversion service with interactive driver. | modules | 7 | 0 |
modules/mysql_hookandroot_lib.c |
reverse_shell: fork & send a bash shell to the attacker before starting mysqld | modules | 6 | 0 |
modules/network_opsec.py |
Network OPSEC — proxy chain enforcement, canary detection, traffic randomization. | modules | 12 | 1 |
modules/nmap2csv.py |
This file is part of nmaptocsv. | modules | 44 | 0 |
modules/notify.py |
Stateless multi-channel notifier for finding events. | modules | 17 | 7 |
modules/nuclei_templates_sync.py |
Sync community Nuclei template collections into the LazyOwn marketplace. | modules | 16 | 0 |
modules/obs_parser.py |
modules/obs_parser.py | modules | 41 | 11 |
modules/ooficesod0woodo.py |
- | modules | 0 | 0 |
modules/operation.py |
Caldera-style operation lifecycle: create, start, pause, resume, stop, status. | modules | 23 | 1 |
modules/operator_profiles.py |
Multi-operator profile management for LazyOwn team server. | modules | 20 | 1 |
modules/opsec_scorer.py |
OPSEC scoring engine — pre-execution noise assessment and gated scoring. | modules | 36 | 4 |
modules/orchestrator_config.py |
Operator configuration for the loop core: pipeline stages and auto verbs. | modules | 13 | 6 |
modules/osint_enrich.py |
Keyless OSINT enrichment for finding evidence. | modules | 13 | 2 |
modules/payload_factory.py |
Native payload generation framework — stagers, stages, singles, formats. | modules | 32 | 6 |
modules/phishing_orchestrator.py |
Smart Phishing Campaign Orchestrator. | modules | 23 | 3 |
modules/pipeline_engine.py |
modules/pipeline_engine.py — Declarative YAML pipelines for LazyOwn | modules | 80 | 5 |
modules/planner.py |
Fact-based planner — decides the next ability to run. | modules | 28 | 8 |
modules/playbook_engine.py |
modules/playbook_engine.py | modules | 31 | 6 |
modules/playbook_executor.py |
Playbook Executor — bridges MITRE ATT&CK playbooks to the pipeline engine. | modules | 25 | 0 |
modules/polymorphic_engine.py |
Polymorphic code generation engine — shellcode mutation and obfuscation. | modules | 19 | 1 |
modules/privesc_predictor.py |
Crystal Ball — privilege escalation vector prediction engine. | modules | 17 | 1 |
modules/professional_report.py |
Professional Red Team Report Generator. | modules | 22 | 2 |
modules/r.sh |
Obtener la versión del kernel actual | modules | 1 | 9 |
modules/reactive_engine.py |
modules/reactive_engine.py | modules | 42 | 7 |
modules/recommender.py |
LazyOwn Smart Command Recommender | modules | 4 | 1 |
modules/redteam_gym.py |
Red Team Gym — gamified pentest training mode. | modules | 18 | 3 |
modules/reflective_dll.py |
Reflective DLL loading — load PE from memory without touching disk. | modules | 27 | 0 |
modules/resource_script.py |
Enhanced resource script engine — variables, conditionals, loops, macros. | modules | 21 | 3 |
modules/reverse-shell.c |
reverse_shell_init: call_usermodehelper function is used to create user mode processes from... | modules | 2 | 0 |
modules/revshell.c |
- | modules | 2 | 0 |
modules/rich_tui.py |
Rich-based live TUI dashboard for LazyOwn campaign monitoring. | modules | 14 | 1 |
modules/rl_trainer.py |
modules/rl_trainer.py | modules | 31 | 5 |
modules/rootkit/mr.c |
- | rootkit | 33 | 0 |
modules/rootkit/mrhyde.c |
original_dirent: define HIDE_DIR ".grisun0" define HIDE_USER "grisun0" define MAX_HIDE_PIDS 140... | rootkit | 38 | 0 |
modules/rootkit/mrhyde2.c |
original_dirent: define HIDE_DIR ".grisun0" define HIDE_USER "grisun0" define MAX_HIDE_PIDS 140... | rootkit | 35 | 0 |
modules/rootkit/mrhyde3.c |
io_uring_sq: #define HIDDEN_FILE9 "listener_{line}.sh" #define PATHMRHYDE "mrhyde.so" #define... | rootkit | 57 | 0 |
modules/rootkit/rootkit.asm |
- | rootkit | 1 | 0 |
modules/rootkit/rootkit.c |
Define the process name and file name to hide | rootkit | 13 | 0 |
modules/saas_attacks.py |
SaaS attacks — Microsoft 365, Google Workspace, Salesforce, ServiceNow exploitation. | modules | 15 | 1 |
modules/scripts/clean_history.sh.sh |
- | misc | 0 | 0 |
modules/search.py |
Autor: Gris Iscomeback Correo electrónico: grisiscomeback[at]gmail[dot]com Fecha de creación... | modules | 0 | 1 |
modules/security_sanitizers.py |
Security sanitizers shared across the LazyOwn framework. | modules | 25 | 5 |
modules/selfupdate.py |
Guarded self-update for the LazyOwn working tree. | modules | 12 | 4 |
modules/session_cleanup.py |
Ephemeral infrastructure cleanup on shell exit. | modules | 5 | 2 |
modules/session_rag.py |
— ChromaDB-backed RAG over sessions/ artefacts. | modules | 24 | 6 |
modules/session_reader.py |
modules/session_reader.py | modules | 29 | 3 |
modules/session_state.py |
LazyOwn Session State | modules | 9 | 4 |
modules/sidequestion.py |
Isolated side-question answering against immutable snapshots. | modules | 9 | 2 |
modules/sleep_obfuscation.py |
Sleep obfuscation engine for beacon memory evasion. | modules | 28 | 3 |
modules/smbver.sh |
Author: rewardone Description: Requires root or enough permissions to use tcpdump Will listen... | modules | 0 | 0 |
modules/socks_proxy.py |
SOCKS5 proxy engine for tunneling traffic through LazyOwn beacons. | modules | 31 | 3 |
modules/staged_delivery.py |
Staged payload delivery — HTA, Office macros, LNK files, ISO/VHD packaging. | modules | 18 | 1 |
modules/state_manager.py |
UnifiedStateManager — single source of truth for all LazyOwn campaign state. | modules | 41 | 7 |
modules/sudo_tiocsti.py |
– Production‑grade TIOCSTI injection with multiple attack modes. | modules | 9 | 0 |
modules/surface_ops.py |
Pure attack-surface helpers (no cmd2): discovery targets, confirm parsing. | modules | 4 | 2 |
modules/tel.py |
to_numbers: Simula la función toNumbers de JavaScript | modules | 6 | 0 |
modules/test_lazyencoder_decoder.py |
- | modules | 0 | 0 |
modules/threat_model.py |
— Blue team threat model builder for LazyOwn. | modules | 16 | 4 |
modules/timeline_narrator.py |
LazyOwn Timeline Narrator | modules | 6 | 3 |
modules/timestomper.py |
Filesystem timestomping — MACB timestamp manipulation for stealth. | modules | 12 | 1 |
modules/tmp.sh |
- | modules | 0 | 0 |
modules/tools/__init__.py |
Reusable development tools for the LazyOwn autonomous loop. | modules_tools | 0 | 1 |
modules/tools/loop_smoke.py |
End-to-end smoke of the shared loop runner (no LLM, no network). | modules_tools | 10 | 0 |
modules/tools/monitor_smoke.py |
Headless smoke of the Textual worker monitor (lazy_view). |
modules_tools | 3 | 0 |
modules/toolset.py |
Typed tool sets per agent role with shell allowlists. | modules | 5 | 1 |
modules/toposwarm_bridge.py |
modules/toposwarm_bridge.py | modules | 22 | 6 |
modules/traffic_morpher.py |
C2 traffic obfuscation: domain fronting, protocol mimicking, and traffic shaping. | modules | 11 | 1 |
modules/traffic_recorder.py |
HTTP traffic recorder with file-tree storage and SQLite index. | modules | 23 | 11 |
modules/ttp_coverage.py |
TTP coverage matrix — real-time MITRE ATT&CK technique tracking. | modules | 13 | 1 |
modules/ttpscope_bridge.py |
Secure bridge for the TTPScope MITRE ATT&CK research client. | modules | 33 | 1 |
modules/unified_bridge.py |
UnifiedBridge — single API over all LazyOwn bridges and routing engines. | modules | 29 | 1 |
modules/unified_dashboard.py |
Unified campaign dashboard — combines all state sources into one view. | modules | 15 | 2 |
modules/update_db.sh |
Nombre del script: update_db.sh Autor: Gris Iscomeback Correo electrónico... | modules | 0 | 0 |
modules/venator.py |
- | modules | 34 | 0 |
modules/vuln_bot_cli.py |
- | modules | 2 | 0 |
modules/vulnbot.py |
list_files: Lista archivos en un directorio | modules | 21 | 1 |
modules/websocket_beacon.py |
WebSocket beacon transport for modern C2 channels. | modules | 21 | 0 |
modules/win_rootkit/backup.c |
- | win_rootkit | 17 | 0 |
modules/win_rootkit/mrhyde.c |
RunExperiment: Define the RunExperiment function | win_rootkit | 18 | 0 |
modules/win_rootkit/win_rin3_rootkit.cs |
- | win_rootkit | 10 | 0 |
modules/win_rootkit/win_ring3_rootkit.c |
initPIDArray: Función para inicializar el arreglo de PIDs | win_rootkit | 27 | 0 |
modules/win_rootkit/win_ring3_rootkit.cpp |
- | win_rootkit | 9 | 0 |
modules/wineconfig.sh |
- | modules | 0 | 0 |
modules/workspace.py |
Confined workspace file browser for operators and loop agents. | modules | 7 | 1 |
modules/world_model.py |
modules/world_model.py | modules | 61 | 32 |
modules/yaml_generator.py |
load_payload: Carga las variables desde payload.json | modules | 8 | 0 |
modules/yara_scanner.py |
YARA integration for malware classification and IOC scanning. | modules | 16 | 2 |
mutate.sh |
- | root | 0 | 0 |
plugins/generate_c_reverse_shell.lua |
- | plugins | 2 | 0 |
plugins/generate_cleanup_commands.lua |
- | plugins | 1 | 0 |
plugins/generate_html_payload.lua |
- | plugins | 1 | 0 |
plugins/generate_lateral_command.lua |
- | plugins | 1 | 0 |
plugins/generate_linux_asm_reverse_shell.lua |
- | plugins | 2 | 0 |
plugins/generate_linux_raw_shellcode.lua |
- | plugins | 1 | 0 |
plugins/generate_lolbird.lua |
- | plugins | 6 | 0 |
plugins/generate_msfvenom_loader.lua |
- | plugins | 5 | 0 |
plugins/generate_msfvenom_loader_windows.lua |
- | plugins | 1 | 0 |
plugins/generate_reverse_shell.lua |
- | plugins | 1 | 0 |
plugins/generate_stub.lua |
- | plugins | 5 | 0 |
plugins/init_plugins.lua |
- | plugins | 1 | 0 |
plugins/kerberos_harvest.lua |
- | plugins | 1 | 0 |
plugins/lolbas_bitsadmin_exe.lua |
- | plugins | 0 | 0 |
plugins/lolbas_certutil_download_exec.lua |
- | plugins | 3 | 0 |
plugins/lolbas_certutil_exe.lua |
- | plugins | 1 | 0 |
plugins/lolbas_mshta_js.lua |
- | plugins | 0 | 0 |
plugins/lolbas_mshta_reverse_shell.lua |
- | plugins | 0 | 0 |
plugins/lolbas_rundll32_dll.lua |
- | plugins | 0 | 0 |
plugins/lolbas_wmic_xsl_execution.lua |
- | plugins | 1 | 0 |
plugins/parse_nmap_with_xmlstarlet.lua |
- | plugins | 1 | 0 |
plugins/run_nuclei_on_nmap_files.lua |
- | plugins | 1 | 0 |
plugins/run_python_rev_c2.lua |
- | plugins | 1 | 0 |
plugins/rundll32_sct_from_url.lua |
- | plugins | 4 | 0 |
plugins/validate_shellcode.lua |
- | plugins | 4 | 0 |
plugins/visualize_network.lua |
- | plugins | 1 | 0 |
poc_tui/__main__.py |
Entry point: python3 -m poc_tui | poc_tui | 0 | 0 |
poc_tui/app.py |
LazyOwn TUI Shell — Proof of Concept. cmd2 as backend, Textual as frontend. | poc_tui | 47 | 3 |
poc_tui/config.py |
Payload.json configuration manager for the TUI shell. | poc_tui | 11 | 1 |
poc_tui/plugin_loader.py |
Unified plugin loader — YAML addons, Lua plugins, and .tool files. | poc_tui | 22 | 0 |
poc_tui/run.py |
Convenience launcher that symlinks LazyOwn dirs and starts the TUI. | poc_tui | 1 | 0 |
poc_tui/test_app.py |
Pytest tests for LazyOwn TUI Shell POC. | poc_tui | 56 | 0 |
pwntomate.py |
This software must not be used by military or secret service organisations. | root | 3 | 0 |
py2elf.sh |
Nombre del script: py2elf.sh Autor: Gris Iscomeback Correo electrónico... | root | 0 | 0 |
readmeneitor.py |
— Automated documentation generator for LazyOwn RedTeam Framework. | root | 11 | 0 |
run_telegram_hermes.sh |
Launcher for telegram_hermes.py Uses the dedicated venv with python-telegram-bot dependencies | root | 0 | 0 |
run_topoexploit_agent.sh |
Run TopoExploit against an ExploitGym task using the local model. | root | 2 | 0 |
scripts/__init__.py |
Repository-side maintenance scripts. | scripts | 0 | 0 |
scripts/activate_migrations.py |
Activate dormant CommandSet migrations. | scripts | 7 | 0 |
scripts/backfill_addon_os_trigger.py |
One-shot backfill: add os and trigger keys to every lazyaddon. |
scripts | 5 | 0 |
scripts/check_contract_manifest.py |
Verify that documented contracts still exist on disk. | scripts | 13 | 1 |
scripts/devtools/command_audit.py |
Audit every registered shell command for dispatch and parser errors. | devtools | 4 | 0 |
scripts/devtools/core_smoke.py |
Smoke check for the public surfaces documented in CORE.md. | devtools | 4 | 0 |
scripts/fix_migrated_classes.py |
- | scripts | 0 | 0 |
scripts/generate_sbom.py |
Generate a minimal CycloneDX SBOM from the pinned requirements files. | scripts | 5 | 0 |
scripts/journal.py |
Read-before-you-write journal over GitHub Discussions. | scripts | 14 | 2 |
scripts/lazy_hermes_ask.sh |
— detached Hermes task runner for LazyOwn. | scripts | 0 | 0 |
scripts/lazy_llamacpp_serve.sh |
— idempotent llama.cpp server manager for LazyOwn. | scripts | 1 | 0 |
scripts/lazy_llamacpp_setup.sh |
— one-time local LLM setup for LazyOwn users. | scripts | 1 | 0 |
scripts/lazy_msf_lab.sh |
— Metasploitable2 smoke lab manager for LazyOwn. | scripts | 0 | 0 |
scripts/lazy_opencode_config.py |
Patch operator agent configs to use the Lazy llama.cpp server. | scripts | 6 | 1 |
scripts/migrate_commandsets.py |
Merge _migrated.py CommandSet methods into clean phase modules. | scripts | 6 | 0 |
scripts/migrate_lazyown.py |
Staged migration script: extract do_* methods from lazyown.py into cli/commands/. | scripts | 6 | 1 |
scripts/mutate.sh |
Mutation gate for LazyOwn. | scripts | 3 | 0 |
scripts/mutation_gate.py |
Consolidated mutation gate for LazyOwn. | scripts | 3 | 0 |
scripts/patch_playbook_atomic_ids.py |
Patch APT playbooks: replace placeholder atomic_ids with real technique_ids. | scripts | 0 | 0 |
scripts/pi_qwen_setup.sh |
idempotent bootstrap for the pi_qwen lazyaddon. | scripts | 0 | 0 |
scripts/publish_wiki.sh |
Publish LazyOwn documentation to the GitHub wiki. | scripts | 8 | 0 |
scripts/read_journal.py |
Print the recent engineering journal before a change is written. | scripts | 2 | 0 |
scripts/setup_hermes_mcp.sh |
— register LazyOwn MCP server in Hermes Agent config Usage: bash scripts/setup_hermes_mcp.sh... | scripts | 0 | 0 |
scripts/smoke_onboarding.sh |
Smoke test for the 5-minute onboarding path. | scripts | 1 | 0 |
scripts/sync_doc_stats.py |
Sync documentation numbers with the live codebase — single source of truth. | scripts | 6 | 0 |
scripts/test_bdd.sh |
Behavior-driven (BDD) suite gate for LazyOwn. | scripts | 2 | 0 |
scripts/top_tier_check.py |
Top-tier hygiene audit for LazyOwn. | scripts | 10 | 0 |
scripts/update_apt_atomic_ids.py |
Update APT playbooks with real Atomic Red Team test IDs. | scripts | 2 | 0 |
scripts/validate_agent_contract.sh |
CI validation of the AGENTS.md branching model and coding standards. | scripts | 5 | 0 |
setup.py |
Compatibility shim for legacy python setup.py invocations. |
root | 0 | 0 |
skills/aci_planner.py |
Autonomous Campaign Intelligence (ACI) Planner | skills | 42 | 2 |
skills/autonomous_daemon.py |
skills/autonomous_daemon.py — LazyOwn Autonomous Execution Daemon | skills | 150 | 15 |
skills/autonomous_replay.py |
Deterministic replay of past autonomous-daemon runs. | skills | 19 | 2 |
skills/claude_md_orchestrator/__init__.py |
Public API for the claude_md_orchestrator skill. | claude_md_orchestrator | 0 | 1 |
skills/claude_md_orchestrator/bdd_agent.py |
Behavior-Driven Development agent. | claude_md_orchestrator | 8 | 0 |
skills/claude_md_orchestrator/boy_scout.py |
Boy Scout refactor pass. | claude_md_orchestrator | 5 | 0 |
skills/claude_md_orchestrator/cicd_agent.py |
CI and CD agent. | claude_md_orchestrator | 7 | 0 |
skills/claude_md_orchestrator/config.py |
Centralized configuration for the claude_md_orchestrator skill. | claude_md_orchestrator | 15 | 10 |
skills/claude_md_orchestrator/documentation_agent.py |
Documentation agent. | claude_md_orchestrator | 7 | 0 |
skills/claude_md_orchestrator/models.py |
Data models for the claude_md_orchestrator skill. | claude_md_orchestrator | 32 | 11 |
skills/claude_md_orchestrator/orchestrator.py |
Orchestrator that wires the agents into a deterministic pipeline. | claude_md_orchestrator | 21 | 1 |
skills/claude_md_orchestrator/parser.py |
Parser that turns a CLAUDE.md into actionable contracts. | claude_md_orchestrator | 13 | 4 |
skills/claude_md_orchestrator/reviewer_agent.py |
Code Reviewer and Quality Assurance agent. | claude_md_orchestrator | 10 | 0 |
skills/claude_md_orchestrator/sdd_agent.py |
Spec-Driven Development agent. | claude_md_orchestrator | 6 | 0 |
skills/claude_md_orchestrator/tdd_agent.py |
Test-Driven Development agent. | claude_md_orchestrator | 7 | 1 |
skills/claude_md_orchestrator/tests/conftest.py |
Pytest bootstrap for the claude_md_orchestrator test suite. | skills_claude_md_orchestrator_tests | 0 | 0 |
skills/claude_md_orchestrator/tests/test_orchestrator.py |
Unit tests for the claude_md_orchestrator package. | skills_claude_md_orchestrator_tests | 20 | 0 |
skills/claude_md_orchestrator/validators.py |
DoD validators for every artifact the orchestrator produces. | claude_md_orchestrator | 21 | 6 |
skills/daemon_control.py |
Operator steering layer for the LazyOwn autonomous daemon. | skills | 27 | 4 |
skills/daemon_health.py |
Daemon health monitor — watchdog for the autonomous daemon and C2 listeners. | skills | 16 | 1 |
skills/heartbeat.py |
LazyOwn Heartbeat | skills | 5 | 0 |
skills/hermes-lazyown/__init__.py |
Hermes-LazyOwn integration package. | hermes-lazyown | 0 | 0 |
skills/hermes-lazyown/claudemd_rules.py |
Dynamic Claude.md rule generator for the Hermes-LazyOwn integration. | hermes-lazyown | 14 | 1 |
skills/hermes-lazyown/config_bridge.py |
Configuration bridge: unifies LazyOwn payload.json, environment variables, and fallback defaults... | hermes-lazyown | 16 | 1 |
skills/hermes-lazyown/constants.py |
Central constants for the Hermes-LazyOwn integration layer. | hermes-lazyown | 15 | 6 |
skills/hermes-lazyown/executor.py |
LazyOwn command executor for the Hermes-LazyOwn integration. | hermes-lazyown | 13 | 1 |
skills/hermes-lazyown/hermes_sync.py |
Hermes-native synchronization layer. | hermes-lazyown | 14 | 1 |
skills/hermes-lazyown/mcp_server.py |
Hermes-native MCP server for LazyOwn. | hermes-lazyown | 28 | 0 |
skills/hermes-lazyown/output_compactor.py |
Phase-aware output compaction for the Hermes-LazyOwn integration. | hermes-lazyown | 20 | 1 |
skills/hive_mind.py |
skills/hive_mind.py — LazyOwn Borg Hive Mind | skills | 107 | 6 |
skills/lazyown_automapper.py |
LazyOwn MCP Auto-Mapper | skills | 23 | 1 |
skills/lazyown_campaign.py |
LazyOwn CampaignStore | skills | 26 | 1 |
skills/lazyown_claudemd.py |
LazyOwn CLAUDE.md Hierarchical Loader (Claude Code style) 4-level hierarchy (highest → lowest... | skills | 8 | 2 |
skills/lazyown_context.py |
LazyOwn Context Compaction Pipeline — 5-layer graduated compression (Claude Code style) Layers: 1. | skills | 11 | 3 |
skills/lazyown_daemon.py |
LazyOwn Unified Daemon | skills | 24 | 0 |
skills/lazyown_facts.py |
LazyOwn FactStore | skills | 79 | 7 |
skills/lazyown_groq_agents.py |
skills/lazyown_groq_agents.py | skills | 40 | 4 |
skills/lazyown_hooks.py |
LazyOwn Hook Pipeline (Claude Code style) 8 lifecycle events with chainable handlers. | skills | 14 | 2 |
skills/lazyown_llm.py |
LazyOwn LLM Bridge — MCP Edition | skills | 21 | 4 |
skills/lazyown_mcp.py |
LazyOwn MCP Server Exposes LazyOwn framework capabilities as MCP tools for Claude Code and... | skills | 142 | 8 |
skills/lazyown_mcp_helpers.py |
LazyOwn MCP helper functions — pure logic, no MCP server imports. | skills | 21 | 4 |
skills/lazyown_mcp_opencode.py |
LazyOwn MCP OpenCode bridge. | skills | 3 | 0 |
skills/lazyown_objective.py |
LazyOwn Objective Store | skills | 36 | 7 |
skills/lazyown_parquet_db.py |
LazyOwn Parquet Knowledge Base | skills | 21 | 8 |
skills/lazyown_permissions.py |
LazyOwn Permission System — Deny-First Rule Evaluation (Claude Code style) Deny rules ALWAYS... | skills | 16 | 2 |
skills/lazyown_policy.py |
LazyOwn Policy Engine | skills | 126 | 9 |
skills/lazyown_session.py |
LazyOwn Append-Only Session Transcripts (Claude Code style) All writes are append-only JSONL. | skills | 14 | 2 |
skills/mcp_generated_tools.py |
Auto-generated MCP tool handlers from command_index.json. | skills | 645 | 0 |
skills/mcp_lazy.py |
Single helper behind every lazy optional import of the MCP server. | skills | 1 | 2 |
skills/mcp_restart.sh |
LazyOwn MCP restart helper Kills ALL lazyown_mcp instances (stdio + SSE) and relaunches the SSE... | skills | 0 | 0 |
skills/mcp_tool_generator.py |
Auto-generate MCP tool handlers from cli/command_index.json. | skills | 9 | 0 |
skills/sessions_watcher.py |
LazyOwn Sessions Watcher | skills | 18 | 1 |
skills/setup.sh |
paths_for_sudo: Collect paths that exist on this system | skills | 5 | 0 |
skills/swan_agent.py |
skills/swan_agent.py — LazyOwn SWAN Agent | skills | 44 | 5 |
skills/tests/__init__.py |
- | skills_tests | 0 | 0 |
skills/tests/test_autonomous_daemon.py |
skills/tests/test_autonomous_daemon.py — Tests for autonomous_daemon.py (SOLID refactoring) All... | skills_tests | 88 | 0 |
skills/tests/test_facts.py |
Tests for lazyown_facts.py — parsers and FactStore. | skills_tests | 37 | 0 |
skills/tests/test_harness_e2e.py |
End-to-end test for the Claude Code-style harness layer. | skills_tests | 10 | 0 |
skills/tests/test_hive_mind.py |
skills/tests/test_hive_mind.py — Tests for hive_mind.py (SOLID refactoring) All file I/O uses... | skills_tests | 95 | 0 |
skills/tests/test_mcp_smoke.py |
Smoke test for the MCP server: verify all 131 tools are registered. | skills_tests | 5 | 0 |
skills/tests/test_objectives.py |
Tests for lazyown_objective.py — ObjectiveStore and SoulUpdater. | skills_tests | 28 | 0 |
skills/tests/test_parquet_db.py |
Tests for lazyown_parquet_db.py — ParquetDB. | skills_tests | 33 | 0 |
skills/toposwarm_autonomous.py |
skills/toposwarm_autonomous.py | skills | 15 | 0 |
skills/unified_orchestrator.py |
Unified orchestrator facade collapsing daemon, hive and SWAN into one surface. | skills | 59 | 2 |
skills/update_knowledge.py |
LazyOwn Knowledge Refresh | skills | 1 | 0 |
slack_c2_bot.py |
- | root | 16 | 0 |
source/conf.py |
Configuration file for the Sphinx documentation builder. | misc | 0 | 0 |
static/js/3d-force-graph-1.73.4.min.js |
Version 1.73.4 3d-force-graph - https://github.com/vasturiano/3d-force-graph... | js | 176 | 0 |
static/js/bootstrap-4.5.2.min.js |
r: ! | js | 7 | 0 |
static/js/bootstrap-5.3.0.bundle.min.js |
W: ! | js | 28 | 0 |
static/js/chart.min.js |
xt: ! | js | 88 | 0 |
static/js/force-graph-1.43.5.min.js |
Version 1.43.5 force-graph - https://github.com/vasturiano/force-graph... | js | 31 | 0 |
static/js/html2pdf.bundle.min.js |
o: 3. | js | 738 | 0 |
static/js/jquery-3.5.1.slim.min.js |
! jQuery v3.5.1 -ajax,-ajax/jsonp,-ajax/load,-ajax/script,-ajax/var/location,-ajax/var/nonce,-aja... | js | 6 | 0 |
static/js/particles.js |
- | js | 11 | 0 |
static/js/popper-2.5.4.min.js |
- | js | 1 | 0 |
static/js/purify-3.0.9.min.js |
! @license DOMPurify 3.0.9 | (c) Cure53 and other contributors | Released under the Apache... | js | 2 | 0 |
static/js/quill-2.0.3.js |
! | js | 124 | 0 |
static/js/select2-4.1.0.min.js |
! | js | 3 | 0 |
static/js/showdown-2.1.0.min.js |
! showdown v 2.1.0 - 21-04-2022 | js | 5 | 0 |
static/js/socket.io-4.0.0.min.js |
as: ! | js | 32 | 0 |
static/js/socket.io-4.3.2.min.js |
as: ! | js | 15 | 0 |
static/js/tippy-6.js |
- | js | 3 | 0 |
static/js/vis-network-9.1.2.min.js |
as: @license vis.js is dual licensed under both 1. | js | 135 | 0 |
static/js/vis-network.min.js |
as: @license vis.js is dual licensed under both 1. | js | 135 | 0 |
static/js/xterm-addon-fit-0.7.0.js |
- | js | 0 | 0 |
static/js/xterm-addon-web-links-0.5.0.js |
- | js | 4 | 0 |
static/js/xterm.js |
- | js | 62 | 0 |
telegram_c2.py |
download_c2: Handle the /download_c2 command to download a file from the C2. | root | 18 | 0 |
telegram_hermes.py |
LazyOwn Telegram Bot with Hermes Gateway Integration This bot bridges Telegram to the full... | root | 29 | 0 |
test_bdd.sh |
- | root | 0 | 0 |
testmeneitor.py |
- | root | 5 | 0 |
tests/__init__.py |
- | tests | 0 | 0 |
tests/conftest.py |
Global pytest isolation hooks. | tests | 10 | 0 |
tests/contract/__init__.py |
- | contract | 0 | 0 |
tests/contract/test_suite_homologation.py |
SDD suite homologation contract for LazyOwn tests. | contract | 20 | 0 |
tests/integration_autonomous_flow.py |
test_autonomous_flow_integration: Simulates a simplified autonomous flow to verify module... | tests | 1 | 0 |
tests/run_mutation_addon_creator.py |
Manual mutation testing runner for the LazyAddon creator contract. | tests | 4 | 0 |
tests/run_mutation_api_authz.py |
Mutation testing runner for core.api_authz and core.logging. | tests | 3 | 0 |
tests/run_mutation_autonomous_ops.py |
Manual mutation runner for the LAZY port contracts. | tests | 4 | 0 |
tests/run_mutation_c2_route_auth.py |
Mutation gate for C2 route-level auth boundaries. | tests | 2 | 0 |
tests/run_mutation_contract_manifest.py |
Mutation gate for the contract manifest drift checker. | tests | 2 | 0 |
tests/run_mutation_infra_report.py |
Mutation runner for the disposable infra and reporting contracts. | tests | 4 | 0 |
tests/run_mutation_killchain.py |
Manual mutation testing runner for the unified kill-chain contract. | tests | 4 | 0 |
tests/run_mutation_llm.py |
Mutation runner for the LLM subsystem contract. | tests | 4 | 0 |
tests/run_mutation_no_shell.py |
Mutation gate for the no-shell execution contract. | tests | 2 | 0 |
tests/run_mutation_opsec.py |
Mutation runner for the consolidated OPSEC scorer contract. | tests | 4 | 0 |
tests/run_mutation_phase1.py |
Mutation testing runner for Phase 1 data-gap closure. | tests | 4 | 0 |
tests/run_mutation_shell_semantics.py |
Mutation gate for shell semantics and colored output. | tests | 2 | 0 |
tests/run_mutation_tests.py |
Manual mutation testing runner for LazyOwn connectivity improvements. | tests | 4 | 0 |
tests/run_mutation_ux_usability.py |
Mutation testing runner for the UX usability contracts. | tests | 4 | 0 |
tests/test_aci_planner.py |
tests/test_aci_planner.py | tests | 95 | 0 |
tests/test_addon_creator.py |
Tests for the LazyAddon creator contract and C2 blueprint. | tests | 86 | 0 |
tests/test_aes_key_propagation.py |
TDD tests for the AES key resolution contract. | tests | 12 | 0 |
tests/test_ai_commands_llm.py |
Behavioural tests for the resurrected ask/groq commands. |
tests | 18 | 0 |
tests/test_api_authz.py |
Tests for core.api_authz — tenant-bound API authorization. |
tests | 48 | 0 |
tests/test_api_key_resolution.py |
API-key resolution contract tests. | tests | 9 | 0 |
tests/test_api_v1.py |
Integration tests for the versioned C2 REST API (/api/v1). |
tests | 13 | 0 |
tests/test_asset_gate.py |
Asset gate + un-nerfed outputs + IntelGraph CSP regression tests. | tests | 16 | 0 |
tests/test_atomic_exploit_bridge.py |
Atomic exploit catalogue: one YAML per exploit, engine administered by files. | tests | 36 | 0 |
tests/test_atomic_loop.py |
Atomic emulation inside the autonomous loop (registry + playbooks). | tests | 17 | 0 |
tests/test_atomic_resolver.py |
MITRE ATT&CK <-> Atomic Red Team translator contract. | tests | 9 | 0 |
tests/test_attack_surface3d.py |
Attack-surface 3D module: surface+intel fusion + embedded pages. | tests | 8 | 0 |
tests/test_attack_surface_addons.py |
tests/test_attack_surface_addons.py BDD: Validate cloud, supply-chain, and web attack-surface... | tests | 76 | 0 |
tests/test_auto_crypto.py |
Tests for cli.auto_crypto — automatic session data encryption. | tests | 22 | 0 |
tests/test_auto_registry.py |
Auto-verb registry: loop-ready automation commands. | tests | 20 | 0 |
tests/test_autonomous_ops.py |
BDD scoped tests for integrated autonomous operations C04-C10. | tests | 17 | 0 |
tests/test_autonomous_replay.py |
Tests for [skills/autonomous_replay.py](https://github.com/grisuno/LazyOwn/blob/146a52570ef792d1ec01ec4d3709b0c159990e47/skills/autonomous_replay.py). |
tests | 18 | 0 |
tests/test_autosuggest.py |
Unit tests for cli.autosuggest. | tests | 41 | 0 |
tests/test_backdoor_random.py |
Backdoor credentials: random like C2 operators, persisted via assign. | tests | 5 | 0 |
tests/test_banner_config.py |
Tests for cli/banner_config.py — registry, settings, renderer, wizard. | tests | 45 | 0 |
tests/test_bdd_infra_range_report.py |
BDD scenarios for disposable infra, range, and executive reporting. | tests | 5 | 0 |
tests/test_beacon_config_builder.py |
Tests for modules/beacon_config_builder.py — BeaconConfig, BeaconConfigBuilder, and... | tests | 16 | 0 |
tests/test_beacon_history.py |
BDD-style tests for persistent beacon command/result history storage. | tests | 10 | 0 |
tests/test_blacksandbeacon_addon.py |
tests/test_blacksandbeacon_addon.py Validation suite for the blacksandbeacon and... | tests | 60 | 0 |
tests/test_bof_registry.py |
Tests for modules/bof_registry.py — BofEntry, BofCatalog, BofRegistry, BofMarketplace... | tests | 51 | 0 |
tests/test_bootstrap_installer.py |
Integration tests for the installer scripts (no network, no apt). | tests | 14 | 0 |
tests/test_boyscout_contracts.py |
Boy-scout contracts for critical shared surfaces. | tests | 15 | 0 |
tests/test_bridge_catalog_filtered.py |
tests/test_bridge_catalog_filtered.py Tests for BridgeDispatcher.catalog_summary_filtered... |
tests | 20 | 0 |
tests/test_c2_agent.py |
Malleable C2 agent: template safety, key ownership, record reading. | tests | 20 | 0 |
tests/test_c2_profile_engine.py |
Tests for modules/c2_profile_engine.py — TlsProfile, DnsProfile, SmbProfile, WebSocketProfile... | tests | 63 | 0 |
tests/test_c2_route_auth.py |
Guard the C2 route-level auth boundaries. | tests | 5 | 0 |
tests/test_c2_shell_lazy.py |
The C2 must not build the CLI shell at module load. | tests | 14 | 0 |
tests/test_c2_surfaces.py |
C2 dashboard surfaces fed by the autonomous loop and APT chains. | tests | 12 | 0 |
tests/test_campaign_import_safety.py |
Campaign import trust boundaries hold against hostile packages. | tests | 14 | 0 |
tests/test_categories.py |
Tests for modules/categories.py — category constants and look-up tables. | tests | 47 | 0 |
tests/test_chain_mode.py |
Tests for cli/chain_mode.py. | tests | 28 | 0 |
tests/test_chain_registry.py |
Chain registry: orchestration from chains/*.yaml, never hardcoded steps. | tests | 9 | 0 |
tests/test_ci_strict.py |
Contract tests for the strict CI workflow. | tests | 10 | 0 |
tests/test_claudemd_consistency.py |
Guard against drift between CLAUDE.md numeric claims and reality. | tests | 3 | 0 |
tests/test_claudemd_size.py |
Enforce a hard cap on [CLAUDE.md](https://github.com/grisuno/LazyOwn/blob/146a52570ef792d1ec01ec4d3709b0c159990e47/CLAUDE.md) size. |
tests | 1 | 0 |
tests/test_cli_assign.py |
do_assign / do_show behaviour and shell wiring tests. |
tests | 39 | 0 |
tests/test_cli_auth_remember.py |
Remember-me token comparison must be timing-safe. | tests | 7 | 0 |
tests/test_cli_command_sets.py |
cli package declarativeness and CommandSet registration tests. |
tests | 56 | 0 |
tests/test_cli_enhancements.py |
Tests for cli/cli_enhancements.py and the audit CommandSet. | tests | 50 | 0 |
tests/test_collab_and_onboarding.py |
tests/test_collab_and_onboarding.py Tests for: - Gap #2 (team server): collab_bp UI route, SSE... | tests | 91 | 0 |
tests/test_command_allowlist.py |
TDD tests for the command allowlist contract. | tests | 14 | 0 |
tests/test_command_allowlist_behavior.py |
BDD behavior scenarios for the command allowlist contract. | tests | 3 | 0 |
tests/test_command_chain.py |
Tests for cli/command_chain.py. | tests | 24 | 0 |
tests/test_command_form.py |
Tests for cli/command_form.py. | tests | 13 | 0 |
tests/test_command_namespace_uniqueness.py |
Cross-CommandSet command-name uniqueness contract. cmd2 refuses to register a second... | tests | 3 | 0 |
tests/test_command_palette.py |
Command palette test suite. | tests | 238 | 0 |
tests/test_command_set_migration.py |
Tests for the completed migration of LazyOwnShell commands. |
tests | 42 | 0 |
tests/test_conditional_hooks_extended.py |
Tests for extended HookEngine trigger matching. | tests | 17 | 0 |
tests/test_console_stdio.py |
Non-blocking stdio hardening: logging must never crash the shell. | tests | 14 | 0 |
tests/test_contract_manifest.py |
Tests for the contract manifest drift checker. | tests | 4 | 0 |
tests/test_core.py |
core package and utils re-export tests. |
tests | 47 | 0 |
tests/test_core_command_bridge.py |
Tests for core.command_bridge — CommandBridge lazy shell facade. | tests | 14 | 0 |
tests/test_core_config.py |
Tests for core.config — Config wrapper, load_payload, save_payload, load_and_validate. | tests | 19 | 0 |
tests/test_core_defaults.py |
Build-time defaults are data, not literals. | tests | 20 | 0 |
tests/test_core_executor.py |
Tests for core.executor — safe_run and run_shell subprocess wrappers. | tests | 15 | 0 |
tests/test_core_modules.py |
tests/test_core_modules.py Integration test suite for LazyOwn core modules. | tests | 94 | 0 |
tests/test_cors_behavior.py |
BDD behavior scenarios for the CORS policy contract. | tests | 4 | 0 |
tests/test_cors_policy.py |
TDD tests for the CORS policy contract. | tests | 17 | 0 |
tests/test_cors_socketio_regression.py |
Regression tests for the Socket.IO origin allowlist. | tests | 8 | 0 |
tests/test_credential_vault.py |
Tests for core/credential_vault.py | tests | 16 | 0 |
tests/test_credentials_rotation.py |
Tests for CredMigratedCommandSet._rotate_existing_credentials. | tests | 9 | 0 |
tests/test_csrf_behavior.py |
BDD behavior scenarios for the CSRF contract. | tests | 6 | 0 |
tests/test_csrf_policy.py |
TDD tests for the CSRF policy contract. | tests | 24 | 0 |
tests/test_daemon_control.py |
Unit tests for skills.daemon_control. | tests | 29 | 0 |
tests/test_daemon_ctl_command_set.py |
Daemon command set extraction tests. | tests | 9 | 0 |
tests/test_daemon_loop.py |
BDD tests for the integrated autonomous loop in the daemon. | tests | 30 | 0 |
tests/test_dashboard_routes.py |
Integration tests for dashboard_bp loot and timeline routes. | tests | 29 | 0 |
tests/test_dashboard_tui.py |
Tests for cli/dashboard_tui.py. | tests | 23 | 0 |
tests/test_db.py |
Tests for modules.db — LazyOwnDB SQLite layer. | tests | 13 | 0 |
tests/test_dependencies.py |
Tests for core.dependencies graceful optional-import handling. |
tests | 35 | 0 |
tests/test_detection_feed.py |
Tests for modules/detection_feed.py. | tests | 19 | 0 |
tests/test_docs_drift_contract.py |
Docs drift workflow contract tests. | tests | 11 | 0 |
tests/test_doctor.py |
tests/test_doctor.py Tests for the preflight environment health check in cli.doctor. |
tests | 34 | 0 |
tests/test_encoding_command_set.py |
Encoding command set extraction tests. | tests | 9 | 0 |
tests/test_engage_gates.py |
BDD tests for gates and transcripts in the live engage path. | tests | 28 | 0 |
tests/test_engage_orchestrator.py |
tests/test_engage_orchestrator.py Coverage for the engage feature added in... | tests | 77 | 0 |
tests/test_engage_summary.py |
Tests for modules/engage_summary.py. | tests | 10 | 0 |
tests/test_engagement_and_ping.py |
tests/test_engagement_and_ping.py | tests | 40 | 0 |
tests/test_engagement_command_gate.py |
Tests for the command-recording gate of cli/engagement_hooks.py. | tests | 18 | 0 |
tests/test_engagement_elo_and_methodology.py |
Tests for the ELO + karma + methodology extensions of cli/engagement_hooks.py. | tests | 43 | 0 |
tests/test_estorides_fusion.py |
Estorides OSINT fusion: load, merge, surface readers. | tests | 12 | 0 |
tests/test_evidence_hints.py |
Tests for evidence-backed inline hints. | tests | 35 | 0 |
tests/test_executor_kernel.py |
Single execution kernel shared by every command runner. | tests | 15 | 0 |
tests/test_exploitgym_gym.py |
Tests for the ExploitGym integration module. | tests | 27 | 0 |
tests/test_exploration_and_addons.py |
tests/test_exploration_and_addons.py Coverage for the new exploration engine and addon schema... | tests | 36 | 0 |
tests/test_finding_lifecycle.py |
Finding confirmation lifecycle + keyless OSINT enrichment. | tests | 10 | 0 |
tests/test_fuzzy_picker.py |
Tests for cli/fuzzy_picker.py and lint-regression for lazyown.py. | tests | 25 | 0 |
tests/test_graph_advisor.py |
Tests for cli/graph_advisor.py. | tests | 25 | 0 |
tests/test_graph_overlay.py |
Tests for cli/graph_overlay.py. | tests | 14 | 0 |
tests/test_gui_xss_sinks.py |
Regression gate for XSS sinks in the C2 dashboard template. | tests | 9 | 0 |
tests/test_hash_cracker.py |
Tests for modules/hash_cracker.py | tests | 22 | 0 |
tests/test_help_ui_command_set.py |
Help/UI command set extraction tests. | tests | 9 | 0 |
tests/test_host_policy.py |
Single host-validation policy shared by CLI gates and hardening. | tests | 9 | 0 |
tests/test_html_sanitizer.py |
TDD tests for the bleach-based HTML sanitizer contract. | tests | 16 | 0 |
tests/test_https_redirect.py |
TDD tests for the HTTPS redirect contract. | tests | 11 | 0 |
tests/test_improvements_spec.py |
Spec-driven validation of the three priority improvements. | tests | 160 | 0 |
tests/test_infra_disposable.py |
TDD unit tests for the disposable infrastructure contracts. | tests | 35 | 0 |
tests/test_input_fuzz.py |
Fuzzing for input validators and rating boundaries. | tests | 8 | 0 |
tests/test_intel_fusion.py |
Intel fusion + LLM defaults regression tests. | tests | 7 | 0 |
tests/test_intelligence_engine.py |
Tests for IntelligenceEngine — collection, analysis, intelligence, dissemination. | tests | 27 | 0 |
tests/test_journal.py |
Tests for the GitHub Discussions engineering journal. | tests | 10 | 0 |
tests/test_killchain.py |
Tests for cli/killchain.py. | tests | 9 | 0 |
tests/test_killchain_auto_refresh.py |
BDD-style tests for the periodic kill-chain auto-refresh in the tips engine. | tests | 8 | 0 |
tests/test_killchain_gap_signal.py |
Tests for :class:cli.recommendation_signals.KillchainGapSignal. |
tests | 20 | 0 |
tests/test_killchain_snapshot.py |
BDD-style tests for the unified kill-chain snapshot contract. | tests | 12 | 0 |
tests/test_killchain_unified.py |
Tests for unified killchain phase: read_phase, write_phase, phase mapping helpers. | tests | 9 | 0 |
tests/test_killchain_unified_v2.py |
Comprehensive tests for modules.killchain — the unified single source of truth. | tests | 42 | 0 |
tests/test_lazy_agent.py |
Agentic tool-calling: worker agent path, resume and main sessions. | tests | 18 | 0 |
tests/test_lazy_engine.py |
Lazy engine (LAZY-faithful) regression tests with stub executor. | tests | 11 | 0 |
tests/test_lazy_evidence.py |
Traffic recorder trigram search and frozen evidence package export. | tests | 3 | 0 |
tests/test_lazy_loop.py |
Production loop fidelity tests: guard judge, findings, settlement, bridge. | tests | 36 | 0 |
tests/test_lazy_orchestrator.py |
Orchestrator admin console: headless smoke and real persistence. | tests | 18 | 0 |
tests/test_lazy_parity.py |
LAZY parity proof tests: every ported subsystem must behave, not just exist. | tests | 26 | 0 |
tests/test_lazy_proxy.py |
Recorder proxy: CA, HTTP/HTTPS capture, engine ingestion, auto-binding. | tests | 28 | 0 |
tests/test_lazy_runner.py |
Shared loop runner, monitor surface and store-fidelity regressions. | tests | 143 | 0 |
tests/test_lazy_streaming.py |
Token-level streaming: factory, pool deltas, agent reassembly. | tests | 17 | 0 |
tests/test_lazy_worker_safety.py |
Worker safety: step sanitizer, task-wide dedupe, sudo skip, worker pool. | tests | 12 | 0 |
tests/test_lazygui_backend.py |
Tests for lazygui backend services. | tests | 59 | 0 |
tests/test_lazygui_graph_widget.py |
Tests for lazygui graph widget contracts. | tests | 51 | 0 |
tests/test_lazygui_models.py |
Tests for lazygui domain models. | tests | 42 | 0 |
tests/test_lazynmap_post.py |
Spec coverage for :mod:cli.lazynmap_post. |
tests | 16 | 0 |
tests/test_lazynmap_sudo_branches.py |
LazyNmap sudo branches: unattended abort vs interactive prompt. | tests | 4 | 0 |
tests/test_legacy_shims.py |
Legacy shim contract plus referenced-script existence. | tests | 5 | 0 |
tests/test_lesson_ingestor.py |
Tests for modules/lesson_ingestor.py. | tests | 11 | 0 |
tests/test_lint_quality.py |
Lint quality gate. | tests | 5 | 0 |
tests/test_live_surface.py |
Tests for modules/live_surface.py. | tests | 10 | 0 |
tests/test_llm_adapter_parity.py |
Parity contract tests for the self-contained Groq adapter family. | tests | 41 | 0 |
tests/test_llm_budget.py |
Contract tests for the LLM budget cap. | tests | 27 | 0 |
tests/test_llm_contract.py |
LLM subsystem contract tests. | tests | 23 | 0 |
tests/test_llm_legacy_isolation.py |
Watchdog: core must not depend on the retired legacy Groq LLM scripts. | tests | 4 | 0 |
tests/test_llm_prompts.py |
Contract tests for :mod:modules.llm_prompts. |
tests | 27 | 0 |
tests/test_logging_config.py |
Tests for modules.logging_config — resilient file logging. |
tests | 5 | 0 |
tests/test_loop_ai_bridge.py |
Loop AI-stack bridge: event bus, oracle, MoE/RL and SWAN wiring. | tests | 14 | 0 |
tests/test_loop_auto_orchestrator.py |
Loop orchestrates every auto verb plus addon/tool triggers. | tests | 18 | 0 |
tests/test_loop_broadcast.py |
Broadcast board timeline regression tests (lazy 0.3.12 parity). | tests | 3 | 0 |
tests/test_loop_cold_overview.py |
Cold debounce + planner overview parity (lazy 0.3.8 coldgraph). | tests | 6 | 0 |
tests/test_loop_force_graph.py |
Loop ForceGraph context regression tests (lazy 0.3.12 fidelity). | tests | 4 | 0 |
tests/test_loop_guard_audit.py |
Guard audit filters + tool_use_id source jump (lazy 0.3.11/0.3.12 parity). | tests | 4 | 0 |
tests/test_loop_intent_delete.py |
Intent delete modes regression tests (lazy 0.3.12 parity). | tests | 4 | 0 |
tests/test_loop_llm_resilience.py |
LLM retry layers, wall-clock settle, stall-guard (lazy 0.3.8/0.3.9 parity). | tests | 18 | 0 |
tests/test_loop_mentions.py |
Lazy mentions multi-type search and scroll pagination. | tests | 2 | 0 |
tests/test_loop_progress.py |
Loop milestone panel: inferred stage states stay honest. | tests | 11 | 0 |
tests/test_loop_scope_traffic_evidence.py |
Scope ownership, port-aware traffic search, role-based evidences (lazy parity). | tests | 7 | 0 |
tests/test_loop_spawn_side.py |
Multi-session chat, persistent side questions, spawn inheritance (lazy 0.3.9). | tests | 4 | 0 |
tests/test_loop_store_race.py |
Store file writes stay atomic under parallel workers (race regression). | tests | 2 | 0 |
tests/test_loop_supervision.py |
Supervision upgrades: repetition watchdog, stack priors, prompt masking. | tests | 9 | 0 |
tests/test_loop_worker_traces.py |
Worker cross-work trace tools (lazy 0.3.10 parity). | tests | 8 | 0 |
tests/test_mcp_dispatch.py |
call_tool dispatches through the handler registry. | tests | 10 | 0 |
tests/test_mcp_improvements.py |
Tests for the high/medium/low impact MCP improvements. | tests | 34 | 0 |
tests/test_mcp_lazy.py |
Contract for the unified MCP lazy-import helper. | tests | 11 | 0 |
tests/test_mcp_recommender_embed.py |
Contract tests for the MCP sitrep recommender embedding. | tests | 8 | 0 |
tests/test_metrics.py |
Tests for [modules/metrics.py](https://github.com/grisuno/LazyOwn/blob/146a52570ef792d1ec01ec4d3709b0c159990e47/modules/metrics.py). |
tests | 14 | 0 |
tests/test_metrics_aware_selector.py |
Tests for MetricsAwareSelector in [skills/autonomous_daemon.py](https://github.com/grisuno/LazyOwn/blob/146a52570ef792d1ec01ec4d3709b0c159990e47/skills/autonomous_daemon.py). |
tests | 24 | 0 |
tests/test_migrate_lazyown_generator.py |
Regression guard for the cli/commands migration generator. |
tests | 6 | 0 |
tests/test_module_registry.py |
Tests for modules.module_registry — ModuleInfo, ModuleRegistry, search, summary. | tests | 29 | 0 |
tests/test_moe_rl_swan.py |
tests/test_moe_rl_swan.py | tests | 59 | 0 |
tests/test_moe_router_check_regression.py |
Regression coverage for modules.moe_router.ExpertAvailabilityChecker._check. |
tests | 6 | 0 |
tests/test_mutation_verification.py |
Mutation verification contracts. | tests | 5 | 0 |
tests/test_nethelpers_command_set.py |
Network helpers command set extraction tests. | tests | 9 | 0 |
tests/test_no_shell_execution.py |
Contract tests: no os.system and no manual sys.stdout reassignment. | tests | 8 | 0 |
tests/test_nuclei_parser.py |
Tests for modules/integrations/nuclei_parser.py | tests | 21 | 0 |
tests/test_ops_loot_phase.py |
Tests for unified loot, cred-graph view, and quantitative phase progress. | tests | 22 | 0 |
tests/test_opsec_scorer.py |
Tests for modules/opsec_scorer.py | tests | 13 | 0 |
tests/test_opsec_scorer_consolidated.py |
Consolidated OPSEC scorer contract tests (SDD/TDD/BDD). | tests | 22 | 0 |
tests/test_packaging.py |
Packaging and project-foundation tests. | tests | 67 | 0 |
tests/test_palette_overlay.py |
Tests for cli/palette_overlay.py. | tests | 10 | 0 |
tests/test_payload_factory.py |
Tests for modules.payload_factory — PayloadFactory, PayloadTemplate, format conversion. | tests | 32 | 0 |
tests/test_payload_schema.py |
Tests for the core.payload_schema validation layer. |
tests | 53 | 0 |
tests/test_phase1_data_gaps.py |
Phase 1 data-gap closure: SDD + TDD + BDD tests. | tests | 55 | 0 |
tests/test_phase_labels.py |
Tests for cli.phase_labels — canonical phase display labels contract. | tests | 5 | 0 |
tests/test_pi_leg.py |
pi loop-leg contract tests. | tests | 15 | 0 |
tests/test_pipeline_engine.py |
tests/test_pipeline_engine.py Coverage for Pillar 3 (declarative pipelines)... | tests | 65 | 0 |
tests/test_placeholder_coverage.py |
Placeholder coverage across every registry that resolves {tokens}. |
tests | 8 | 0 |
tests/test_plugin_loader_safety.py |
Marketplace placeholder substitution must be shell-safe. | tests | 7 | 0 |
tests/test_plugin_tiers.py |
Tests for marketplace plugin tiers and operator ratings. | tests | 12 | 0 |
tests/test_profiles.py |
Tests for runtime install profiles (core.profiles). |
tests | 11 | 0 |
tests/test_prompt_readline_markers.py |
Regression tests for readline marker handling in the Neon Box prompt. | tests | 8 | 0 |
tests/test_prompt_refresh.py |
Contract tests for live prompt refresh after a payload change. | tests | 9 | 0 |
tests/test_pwntomate_tools.py |
Regression guard for pwntomate tool templates (tools/*.tool). | tests | 8 | 0 |
tests/test_rea_mcp.py |
Tests for the lazyown_rea MCP tool (REA reverse-engineering skill). | tests | 24 | 0 |
tests/test_reactive_engine_semantic.py |
Tests for the semantic context advisor in [modules/reactive_engine.py](https://github.com/grisuno/LazyOwn/blob/146a52570ef792d1ec01ec4d3709b0c159990e47/modules/reactive_engine.py). |
tests | 13 | 0 |
tests/test_reactive_hints.py |
Tests for cli/reactive_hints.py. | tests | 31 | 0 |
tests/test_reactive_hints_expanded.py |
Tests for cli.reactive_hints — expanded kill-chain tables. | tests | 27 | 0 |
tests/test_reactive_lateral_data.py |
Tests for new reactive engine matchers. | tests | 26 | 0 |
tests/test_reasoning_stream.py |
Tests for cli/reasoning_stream.py. | tests | 13 | 0 |
tests/test_recommendation.py |
Behaviour tests for the unified recommendation engine and its signals. | tests | 36 | 0 |
tests/test_recon_plan.py |
Spec coverage for :mod:cli.recon_plan. |
tests | 34 | 0 |
tests/test_report_banners_endpoints.py |
Tests for /report and /banners endpoints error handling. | tests | 32 | 0 |
tests/test_require_guards.py |
Single-call requirement guards for CommandSet methods. | tests | 5 | 0 |
tests/test_resource_script.py |
Tests for modules/resource_script.py — ScriptContext and ResourceScriptEngine. | tests | 32 | 0 |
tests/test_safe_console.py |
Terminal-safe console: colors and glyphs degrade, never leak. | tests | 4 | 0 |
tests/test_safe_subprocess.py |
TDD tests for the safe subprocess runner contract. | tests | 11 | 0 |
Next: INDEX_p3.md
Generated offline by readmenator from commit 146a52570ef7. Edits here are overwritten by readmenator . --rebuild.
Start
Wiki
- community_0_cli_commands
- community_10_modules_opsec_scorer
- community_11_modules_finding_lifecycle
- community_12_modules_kerberos_core
- community_13_modules_saas_attacks
- community_14_modules_polymorphic_engine
- community_15_contrib_legacy
- community_16_scripts_journal
- community_17_modules_backdoor
- community_18_poc_tui
- community_19_scripts_check_contract_manifest
- community_1_modules_lazyc2
- community_20_scripts_migrate_lazyown
- community_21_tools
- community_22_orphans
- community_2_cli
- community_3_modules_autonomous_daemon
- community_4_modules_lazy_loop
- community_5_lazygui_panels
- community_6_skills_claude_md_orchestrator
- community_7_static_js
- community_8_modules_llm_budget
- community_9_modules_atomic_exploit_bridge
- concepts
- queries
- REPORT
Agent Docs
- API
- API_p10
- API_p11
- API_p12
- API_p13
- API_p14
- API_p15
- API_p16
- API_p17
- API_p18
- API_p19
- API_p2
- API_p20
- API_p21
- API_p22
- API_p23
- API_p3
- API_p4
- API_p5
- API_p6
- API_p7
- API_p8
- API_p9
- ARCHITECTURE
- ARCHITECTURE_p2
- ARCHITECTURE_p3
- ARCHITECTURE_p4
- ARCHITECTURE_p5
- ARCHITECTURE_p6
- ARCHITECTURE_p7
- CONCEPTS
- GOTCHAS
- INDEX
- INDEX_p2
- INDEX_p3
- KB_backdoor
- KB_blueprints
- KB_cgi-bin
- KB_claude_md_orchestrator
- KB_cli
- KB_cli_p2
- KB_cli_p3
- KB_cli_p4
- KB_cli_p5
- KB_commands
- KB_commands_p2
- KB_commands_p3
- KB_commands_p4
- KB_config
- KB_contract
- KB_core
- KB_core_p2
- KB_devtools
- KB_extensions
- KB_hermes-lazyown
- KB_integrations
- KB_js
- KB_js_p2
- KB_js_p3
- KB_js_p4
- KB_lazyc2
- KB_lazygui
- KB_lazyown-docker
- KB_legacy
- KB_legacy_p2
- KB_misc
- KB_modules
- KB_modules_p10
- KB_modules_p11
- KB_modules_p12
- KB_modules_p13
- KB_modules_p14
- KB_modules_p2
- KB_modules_p3
- KB_modules_p4
- KB_modules_p5
- KB_modules_p6
- KB_modules_p7
- KB_modules_p8
- KB_modules_p9
- KB_modules_tools
- KB_palettes
- KB_panels
- KB_plugins
- KB_poc_tui
- KB_root
- KB_root_p2
- KB_rootkit
- KB_scripts
- KB_security
- KB_services
- KB_skills
- KB_skills_claude_md_orchestrator_tests
- KB_skills_p2
- KB_skills_p3
- KB_skills_p4
- KB_skills_p5
- KB_skills_tests
- KB_tests
- KB_tests_p10
- KB_tests_p11
- KB_tests_p12
- KB_tests_p13
- KB_tests_p14
- KB_tests_p15
- KB_tests_p16
- KB_tests_p17
- KB_tests_p2
- KB_tests_p3
- KB_tests_p4
- KB_tests_p5
- KB_tests_p6
- KB_tests_p7
- KB_tests_p8
- KB_tests_p9
- KB_theme
- KB_tools
- KB_widgets
- KB_win_rootkit
- KB_windows
- MEMORY
- SECURITY
- SYMBOLS
- SYMBOLS_p10
- SYMBOLS_p11
- SYMBOLS_p12
- SYMBOLS_p13
- SYMBOLS_p14
- SYMBOLS_p15
- SYMBOLS_p16
- SYMBOLS_p17
- SYMBOLS_p18
- SYMBOLS_p19
- SYMBOLS_p2
- SYMBOLS_p20
- SYMBOLS_p21
- SYMBOLS_p22
- SYMBOLS_p23
- SYMBOLS_p24
- SYMBOLS_p25
- SYMBOLS_p26
- SYMBOLS_p27
- SYMBOLS_p28
- SYMBOLS_p29
- SYMBOLS_p3
- SYMBOLS_p30
- SYMBOLS_p31
- SYMBOLS_p32
- SYMBOLS_p33
- SYMBOLS_p34
- SYMBOLS_p35
- SYMBOLS_p36
- SYMBOLS_p37
- SYMBOLS_p38
- SYMBOLS_p39
- SYMBOLS_p4
- SYMBOLS_p40
- SYMBOLS_p41
- SYMBOLS_p5
- SYMBOLS_p6
- SYMBOLS_p7
- SYMBOLS_p8
- SYMBOLS_p9
Recipes