System Source Map

Developer table mapping every major BreachPilot system to its primary implementation files and documentation links. All implementation paths are repo-relative (relative to the BreachPilot repo root). Documentation links are relative to this file (docs/reference/).

Path conventions

# Every implementation path below was verified to exist with ls/stat
# before this file was written. Paths never include machine prefixes.
ls main.py app.py cli.py db.py tools/api/routes tools/run_service

Rows marked with an Implementation note flag claims that rest on directory layout and doc frontmatter rather than a full read of the implementation.

Entrypoints and CLI dispatch

System	Primary implementation	Documentation
WebUI daemon default launch, direct recon/attack runs, `--menu`, doctor, eval, benchmark flows	`main.py`, `app.py`	architecture.md, runtime-flows.md, cli-reference.md
Legacy research CLI (Flow B, frozen)	`cli.py`, `legacy/cli.py`	components/root/cli.md, runtime-flows.md
Legacy agent loop shims (Flow B)	`agent_loop.py`, `legacy/agent_loop.py`	components/flow-b/agent-loop.md
CLI dispatch layer (config load, exploit settings, recon assessment, safety review, skills flags, interactive menu)	`tools/config_cli.py`, `tools/cli_exploit_settings.py`, `tools/recon_assessment_cli.py`, `tools/safety_review_cli.py`, `tools/skills_cli.py`, `tools/interactive_menu.py`, `tools/attack_ui.py`	cli-reference.md, components/root/main.md
MCP session boot (`open_exploit_mcp_session`)	`tools/mcp_session.py`	mcp/overview.md, architecture.md
Single-target session wiring (ScopeGate + MCP session + agent)	`tools/exploit_session.py`	exploit-agent.md, runtime-flows.md

Implementation note: the exact flag-to-function mapping in main.py (async_main) was not re-read line by line for this table; the entry roles above follow docs/architecture.md and CLAUDE.md.

Configuration

System	Primary implementation	Documentation
Canonical config schema, validator, loader	`tools/config/schema.py`, `tools/config/validator.py`, `tools/config/loader.py`, `tools/config/__init__.py`	configuration/overview.md, config-reference.md
Checked-in lab config and mission scope	`config.yaml`, `mission.yaml`	configuration/environment.md, configuration/secrets.md
Legacy config manager shim	`tools/config_manager.py`	configuration/validation.md
Mission model and risk profiles	`mission.py`	components/flow-b/db-mission.md, safety-model.md

API routes and run-service

System	Primary implementation	Documentation
Run lifecycle endpoints	`tools/api/routes/runs.py`, `tools/api/routes/decisions.py`, `tools/api/routes/events.py`	api/endpoints/runs.md, api/endpoints/decisions.md, api/endpoints/events.md
System, ops, users, connections, benchmarks endpoints	`tools/api/routes/system.py`, `tools/api/routes/ops.py`, `tools/api/routes/users.py`, `tools/api/routes/connections.py`, `tools/api/routes/benchmarks.py`	api/endpoints/system.md, api/overview.md
Evidence graph endpoints	`tools/api/routes/graph.py`, `tools/api/routes/graph_explorer.py`	api/endpoints/graph.md
Route package wiring	`tools/api/routes/__init__.py`	api/overview.md
Transport-neutral run engine (prepare + execute, typed models, providers, tasks)	`tools/run_service/service.py`, `tools/run_service/models.py`, `tools/run_service/providers.py`, `tools/run_service/tasks.py`, `tools/run_service/prepare.py`, `tools/run_service/execute.py`, `tools/run_service/warmup.py`	run-service.md, api/run-manager.md
API daemon auth, errors, session titling, graph service	`tools/api/auth.py`, `tools/api/errors.py`, `tools/api/session_titler.py`, `tools/api/graph_service.py`, `tools/api/graph_builder.py`	api/auth.md, api/overview.md

MCP servers, registry, and tool families

System	Primary implementation	Documentation
Defensive MCP server (scope-gated Nmap/intel)	`mcp_server.py`	mcp/servers/defensive.md, mcp/overview.md
Exploit MCP server (permissive tool surface)	`mcp_exploit_server.py`	mcp/servers/exploit.md, mcp/overview.md
Engine MCP server (read-only advisory)	`mcp_engine_server.py`	mcp/servers/engine.md
Central tool wiring, audit decorator, allowlist gate	`tools/mcp_tools/registry.py`	mcp/overview.md, mcp/registration.md
Shared transport hardening, workspace and redaction helpers	`tools/mcp_shared.py`, `tools/exceptions.py`	mcp/overview.md, mcp/security.md
Terminal execution and target-IP lock	`tools/mcp_tools/terminal/`, `tools/mcp_tools/sandbox_exec.py`	mcp/tool-families/terminal.md
Recon, web-scan, and research families	`tools/mcp_tools/recon.py`, `tools/mcp_tools/web_scan.py`, `tools/mcp_tools/research.py`, `tools/mcp_tools/domain.py`	mcp/tool-families/recon.md, mcp/tool-families/web-scan.md, mcp/tool-families/research.md, mcp/tool-families/domain.md
Attack-module, payload, Metasploit, and cracking families	`tools/mcp_tools/attack_modules.py`, `tools/mcp_tools/payloads.py`, `tools/mcp_tools/metasploit.py`, `tools/mcp_tools/cracking.py`, `tools/mcp_tools/poc_verifier.py`	mcp/tool-families/attack-modules.md, mcp/tool-families/payloads.md, mcp/tool-families/metasploit.md, mcp/tool-families/cracking.md, mcp/tool-families/poc-verifier.md
Sessions, credentials, workspace, and operator families	`tools/mcp_tools/sessions.py`, `tools/mcp_tools/credentials.py`, `tools/mcp_tools/workspace.py`, `tools/mcp_tools/operator_connection.py`, `tools/mcp_tools/ad.py`	mcp/tool-families/sessions.md, mcp/tool-families/credentials.md, mcp/tool-families/workspace.md, mcp/tool-families/ad.md
Agent-assist families (parallel agents, peer models, runtime skills, MITRE, replay)	`tools/mcp_tools/parallel_agents.py`, `tools/mcp_tools/peer_models.py`, `tools/mcp_tools/runtime_skills.py`, `tools/mcp_tools/mitre.py`, `tools/mcp_tools/replay_simulator.py`	mcp/tool-families/parallel-agents.md, mcp/tool-families/peer-models.md, mcp/tool-families/runtime-skills.md, mcp/tool-families/mitre.md, mcp/tool-families/replay-simulator.md
Verification-state families (propose/verify/retest, assessment state, snapshots, killchain)	`tools/mcp_tools/hitl.py`, `tools/mcp_tools/verify.py`, `tools/mcp_tools/retest.py`, `tools/mcp_tools/assessment_state.py`, `tools/mcp_tools/snapshots.py`, `tools/mcp_tools/killchain.py`, `tools/mcp_tools/browser.py`, `tools/mcp_tools/egress_guard.py`	mcp/tool-families/verify.md, mcp/tool-families/retest.md, mcp/tool-families/assessment-state.md, mcp/lifecycle.md

Attack modules and campaign

System	Primary implementation	Documentation
Attack module base, ranking registry, module families	`tools/attack_modules/base.py`, `tools/attack_modules/registry.py`, `tools/attack_modules/modules/`, `tools/attack_modules/graph.py`, `tools/attack_modules/artifacts.py`	attack-modules.md, components/tools/attack-modules/overview.md, components/tools/attack-modules/registry.md
Autonomous campaign engine (state, executor, phases, batch, preflight, persistence)	`tools/campaign/orchestrator.py`, `tools/campaign/state.py`, `tools/campaign/executor.py`, `tools/campaign/phases.py`, `tools/campaign/batch.py`, `tools/campaign/preflight.py`, `tools/campaign/state_store.py`, `tools/campaign/service_tasks.py`, `tools/autonomous_orchestrator.py`	architecture.md, runtime-flows.md
Attack planner, payload crafter, exploit mutator	`tools/attack_planner.py`, `tools/payload_crafter.py`, `tools/exploit_mutator.py`, `tools/exploit_search.py`	attack-modules.md

Exploit agent runner, policy, and swarm

System	Primary implementation	Documentation
Canonical agent loop	`tools/exploit_agent/runner/_impl.py`, `tools/exploit_agent/runner/loop.py`, `tools/exploit_agent/__init__.py`	exploit-agent.md, components/tools/exploit-agent/loop.md, components/tools/exploit-agent/overview.md
Agent policy (permission modes, audit chain)	`tools/exploit_agent/policy.py`	components/tools/exploit-agent/policy.md, safety-model.md
Agent context, prompts, tool calls, reflection, skills	`tools/exploit_agent/context.py`, `tools/exploit_agent/prompt.py`, `tools/exploit_agent/prompts.md`, `tools/exploit_agent/tool_calls.py`, `tools/exploit_agent/tool_catalog.py`, `tools/exploit_agent/reflection.py`, `tools/exploit_agent/skills.py`, `tools/exploit_agent/model_client.py`, `tools/exploit_agent/runner/`	components/tools/exploit-agent/context.md, components/tools/exploit-agent/prompts.md
Swarm orchestrator, blackboard, negotiation, milestones	`tools/swarm/orchestrator.py`, `tools/swarm/blackboard.py`, `tools/swarm/negotiation.py`, `tools/swarm/milestones.py`, `tools/swarm/state_store.py`, `tools/swarm/reflection_run.py`, `tools/swarm/base.py`, `tools/swarm_bridge.py`	swarm.md, components/tools/swarm/overview.md, components/tools/swarm/agents.md
Specialist swarm agents	`tools/swarm/agents/`	components/tools/swarm/agents.md

Skills pipeline, recon, and intelligence

System	Primary implementation	Documentation
Skills pipeline (registry, selection, embeddings, injection, feedback)	`tools/skill_pipeline.py`, `tools/skill_registry.py`, `tools/skill_registry_cache.py`, `tools/skill_selector.py`, `tools/skill_embeddings.py`, `tools/skill_feedback.py`, `tools/skills_cli.py`, `tools/skill_author.py`, `skills/`	skills.md, skill-authoring.md
Recon pipeline (scanner, enumerator, config)	`tools/recon/pipeline.py`, `tools/recon/scanner.py`, `tools/recon/enumerator.py`, `tools/recon/config.py`, `tools/recon_pipeline.py`, `tools/fast_recon.py`, `tools/recon_enrichers.py`, `tools/recon_osint.py`	components/tools/recon/overview.md
Intelligence layer (adapters, belief, evidence, fingerprint, graph, schemas)	`tools/intelligence/`, `tools/cve_lookup.py`, `tools/threat_intel.py`, `tools/web_researcher.py`	components/tools/intelligence/overview.md

Implementation note: the recon/intelligence row file roles follow directory layout and docs/architecture.md; per-module responsibilities were not individually re-read for this table.

Kernel, sandbox, and safety

System	Primary implementation	Documentation
Kernel (audit chain, allowlist, orchestration vocabulary, workspace, config)	`tools/kernel/audit.py`, `tools/kernel/allowlist.py`, `tools/kernel/orchestration.py`, `tools/kernel/workspace.py`, `tools/kernel/config.py`, `tools/kernel/parse.py`, `tools/kernel/discovered.py`	components/tools/kernel/overview.md, safety-model.md
Sandbox (manager, Docker backend/lifecycle, network, policy)	`tools/sandbox/manager.py`, `tools/sandbox/docker_backend.py`, `tools/sandbox/docker_lifecycle.py`, `tools/sandbox/network.py`, `tools/sandbox/policy.py`, `tools/sandbox/mcp_bridge.py`, `tools/sandbox/models.py`	sandbox.md, safety-model.md
Scope, safety review, killchain, snapshots, verification	`scope_gate.py`, `tools/safety_reviewer.py`, `tools/killchain/`, `tools/snapshots.py`, `tools/verification/`, `tools/command_analyzer.py`, `tools/validation_utils.py`	safety-model.md

Benchmark, providers, and plugins

System	Primary implementation	Documentation
Benchmark service (runner, registry, metrics, replay, report, storage)	`tools/benchmark/runner.py`, `tools/benchmark/registry.py`, `tools/benchmark/service.py`, `tools/benchmark/metrics.py`, `tools/benchmark/replay.py`, `tools/benchmark/report.py`, `tools/benchmark/storage.py`, `tools/benchmark_cli.py`, `benchmarks/xben/`	benchmarks.md
Graded eval harness	`tools/eval_harness.py`, `tools/eval_benchmark.py`, `tools/eval_checks.py`, `eval_targets/`	evaluation.md
Provider registry and model routing	`tools/providers/registry.py`, `tools/providers/base.py`, `tools/providers/types.py`, `tools/providers/ollama_provider.py`, `tools/providers/opencode_go_provider.py`, `tools/providers/chatgpt_provider.py`, `tools/model_router.py`, `tools/ollama_models.py`	providers.md, provider-development.md
Plugin system and bundled plugins	`tools/plugins.py`, `plugins/`	plugin-development.md, extension-guide.md

Operator connection and credential vault

System	Primary implementation	Documentation
Operator connection (manager, implants)	`tools/operator_connection/manager.py`, `tools/operator_connection/implants.py`, `tools/mcp_tools/operator_connection.py`	mcp/overview.md
Encrypted credential vault and store-backed MCP tools	`tools/credential_store.py`, `tools/mcp_tools/credentials.py`, `tools/api_key_store.py`	mcp/tool-families/credentials.md, configuration/secrets.md

Implementation note: the operator-connection doc link points at the MCP overview because no dedicated operator-connection doc was found in docs/; the implementation paths above were verified to exist.

Logs, database, and persistence

System	Primary implementation	Documentation
Decision log and run log	`tools/decision_log.py`, `tools/run_log.py`, `tools/activity_log.py`	api/persistence.md, outcome-evidence.md
Flow B SQLite schema (missions, tasks, evidence, findings, audit)	`db.py`, `memory.py`, `evidence.py`	database-mission.md, components/flow-b/db-mission.md, components/flow-b/evidence-memory-graph.md
API persistence, run manager, event and decision brokers	`tools/api/persistence.py`, `tools/api/run_manager.py`, `tools/api/event_broker.py`, `tools/api/decision_broker.py`	api/persistence.md, api/event-broker.md, api/run-manager.md
Sessions, resume, goal and experience state	`tools/session_manager.py`, `tools/persistent_session_manager.py`, `tools/resume_state.py`, `tools/goal_engine.py`, `tools/experience_store.py`, `tools/semantic_memory.py`, `tools/attack_memory.py`	api/persistence.md

WebUI

System	Primary implementation	Documentation
Routes and pages	`webui/src/routes/`, `webui/src/App.tsx`, `webui/src/main.tsx`	webui/overview.md, webui/pages/graph.md, webui/pages/home.md, webui/pages/run.md
REST client, generated types, query hooks	`webui/src/api/client.ts`, `webui/src/api/types.ts`, `webui/src/api/hooks.ts`, `webui/src/api/queryClient.ts`	webui/api-integration.md, webui.md
Realtime (WebSocket, SSE, event buffer/store)	`webui/src/api/ws.ts`, `webui/src/api/sse.ts`, `webui/src/api/eventBuffer.ts`, `webui/src/api/eventStore.ts`	api/websocket.md, api/event-broker.md, webui/state.md
Components and feature slices	`webui/src/components/`, `webui/src/features/`	webui/components/graph.md, webui/components/run-create.md

Implementation note: the route-to-doc mapping follows the webui/ doc tree; individual .tsx page responsibilities were not re-read for this table.

Tests, benchmarks, and evaluation

System	Primary implementation	Documentation
Mocked pytest suite (scope, safety, recon, MCP, API, swarm, skills)	`tests/`, `conftest.py`, `tests/test_scope_gate.py`	testing-guide.md, type-checking.md
Oracle benchmark suites	`benchmarks/xben/`, `tools/benchmark/`	benchmarks.md
Eval oracle targets and graded harness	`eval_targets/`, `tools/eval_harness.py`	evaluation.md

architecture.md — system shape, entry points, Flow A/B split
runtime-flows.md — mission creation and agent loop flows
run-service.md — transport-neutral run engine and API layer
api/overview.md — WebUI API lifecycle, auth, brokers
mcp/overview.md — MCP server topology and security model
exploit-agent.md — canonical agent loop reference
swarm.md — multi-agent execution engine
attack-modules.md — attack module system
skills.md — runtime skills system
sandbox.md — disposable execution boundary
safety-model.md — permission model and safety boundaries
providers.md — LLM provider abstraction
webui/overview.md — WebUI architecture
benchmarks.md — reproducible benchmark suites
evaluation.md — graded eval harness
cli-reference.md — CLI flag reference
config-reference.md — configuration reference
glossary.md — subsystem vocabulary

source: repo docs (build sync)Edit this page on GitHub →