Skip to content

Source Evidence and Project Map

English | 中文

06SOURCE EVIDENCE

Every conclusion should return to source, tests, or an explicit limitation

This page is not a file inventory. It explains how the bluebook covers the project, what evidence supports eight judgments, and which questions the repository still cannot prove.

Pass 1

Build a capability map from architecture and every first-party package-group README.

Pass 2

Enter source and tests through the Agent path, governance path, and failure boundaries.

Falsify

Actively seek POC, non-security-boundary, non-recovery, and protocol limitations.

Evidence method

The first pass reads the architecture overview, package-group map, and every first-party package-group README under packages/ to answer which capability families exist and how they compose. The second pass follows Agent creation, model requests, tool execution, session persistence, client projection, subagents, workflows, sandboxing, and protocol entry points into core source and tests to answer how a real task runs and fails.

vendor/ is a manifest-synchronized Cordis upstream copy rather than an independent source for product-innovation claims. native/ and python/ enter coverage through their READMEs and consuming paths. Business value is inferred only where implemented mechanisms support it; market benefit, production scale, and industry completeness are not guessed from code.

Project capability map

Project areaPrimary responsibilityBluebook location
core/Sessions, prompts, tools, Agents, and the agent loopHow one task runs
api/ and typert/Remote BFF, type graph, loader, and RPC registryTyped capability shared across interfaces
llm/LLM Service Definition, catalog, adapters, and DeepSeek providersCapability seams and model configuration
shell/, subprocess/, terminal/Commands, process trees, and persistent terminal executionTool execution and environment constraints
fs/ and lsp/Filesystem capability, policy, and language serversEngineering scenarios and file governance
skill/ and web/Skill registration, catalogs, loaders, search, and fetchKnowledge-work scenarios
context/ and compaction/Request context and model-history compactionLong sessions and replayable context
subagent/ and workflow/Delegation, continuable children, and dynamic workflowsLong-running collaboration and limitations
todo/, plan/, goal/, jobs/, schedule/Plan state, goals, background jobs, and scheduled workObservable task progress
preset/, bundle/, boot/Per-session composition, installable bundles, and application startupEverything is a plugin and platform assembly
guard/ and interaction/Loop hygiene, timeouts, commands, permission, approval, and questionsTool governance and human control
session/Persistence, projections, titles, telemetry, and queriesRecovery, replay, and evidence
settings/, credentials/, identity/User settings, credential references, and anonymous identityConfiguration governance and secret separation
client/ and host/Browser runtime, UI modules, Host, and API proxyPlugin-defined business UI
acp/, sdk/, and python/Automation protocol, JSON-RPC, and two SDKsMultiple access paths
hooks/ and extensions/External-tool bridges and runtime self-modificationEcosystem integration and trusted dynamic extension
sandbox/, e2b/, code-runtime/File confinement, remote-execution POC, and script runtimeGovernance boundaries and current limitations
examples/, support/, util/Runnable compositions, test infrastructure, and zero-dependency utilitiesComposition verification and engineering support

Evidence index for eight innovations

ConclusionPrimary source or documentationRepresentative test evidenceConfidence and limitation
Everything is a pluginArchitecture overview, bundle group, application bootPreset mount testsHigh; plugin trust follows deployment composition
Complete capability seamsCapability seam rules, LLM group, E2B POCLocal sandbox testsHigh; provider security properties differ
Model-visible state is replayableSession implementation, agent loopRequest reconstruction tests, session property testsHigh; compliance retention is outside the log
Tool governance pipelineTool registry, approval serviceTool behavior tests, approval testsHigh; covers actions dispatched through the registry
Per-session Agent compositionAgent preset implementationSession preset tests, discovery testsHigh; this is not an enterprise tenant system
Observable long-running collaborationSubagent group, workflow groupContinuation tests, workflow integration testsHigh; dynamic workflows do not recover after restart
Multiple access pathsACP, SDK, PythonACP turn testsHigh; each protocol exposes its own subset
Client plugin graphClient modules, Web runtimeModule loader tests, conversation testsHigh; domain UX still requires product design

Locate the critical runtime paths

Agent and model requests

The agent loop records turn/start, step/start, user messages, streaming assistant content, tool calls, and end reasons, and uses deriveMessages() for the next request. The session implementation owns append, surface projection, format loading, forks, and message derivation.

Tools and approval

The tool registry declares guards, execution phases, results, and presentation intent. The approval service declares ask/never, fail-closed behavior, and replayable audit events. Together they prove that “the model proposed a call” and “the action was authorized to execute” are separate stages.

Session composition and recovery

Agent presets own discovery, resolution, standing mounts, recomposition limits, and scoped service reads. Session persistence supplies write, read, and preparation paths; continuable subagents preserve their descriptor and delegated policy in their own sessions.

Interface and protocol

The Web runtime builds conversation and trajectory projections from session windows. ACP publishes a narrower automation interface and emits only committed assistant text. SDKs drive the same Agent and session services over JSON-RPC.

Use falsification to bound conclusions

If this fact were foundConclusion it would overturn or narrowCurrent check
Model-visible content had no corresponding session event“Model-visible state is replayable”Main request path records first; no exception found
A tool could reopen execution after a guard denied it“Governance tightens monotonically”ToolGuard has no allow branch
A child Agent could prompt for escalation unattended“Delegated authority is fixed”Delegation appends approval/policy: never
A dynamic workflow resumed after restartThe current non-recovery limitationREADME explicitly states no journal or resume
SandboxMode restricted network and credentials tooThe file-effects-only limitationService vocabulary explicitly excludes network, process, syscall, device, and credentials

What the current repository cannot prove

  • It cannot prove that any industry has a complete out-of-box solution; the repository supplies general capabilities and example compositions.
  • It cannot prove production scale, availability targets, or unit economics; the target deployment needs capacity tests and billing data.
  • It cannot prove session logs automatically meet a specific regulation; compliance depends on identity, storage, and policy.
  • It cannot prove arbitrary model-written code is safe to execute; several worker/VM implementations explicitly are not hostile-code security boundaries.
  • It cannot prove multi-Agent execution is faster or better for every task; benefits depend on decomposability, synthesis, and resource cost.
HOW TO VERIFY NEXT

Enter the main path from a business conclusion, then read its limitations and tests

Beginners start with “From Task to Evidence.” Architecture and platform readers use this index to enter owner packages. Security reviewers work backward from “Governance Boundaries.” READMEs explain obligations, source confirms implementation, and tests prove critical branches remain exercised.

Evidence status: Confirmed against source, tests, and documentation in the current worktree unless noted. No real DeepSeek API, production load, or target-enterprise system validation was run.