Agent System

Grida Agent System

This skill is for the two packages behind Grida Desktop and future hosts (issue #927): @grida/daemon — the local host layer (DaemonServer lifecycle, the loopback HTTP perimeter, files/recents/workspaces, the secrets store, daemon discovery, the tenant seam) — and @grida/agent — the agent TENANT mounted on it (sessions, providers, tool execution, prompt composition, skills, the AI-SDK stream contract). Dependency direction is one-way: @grida/agent imports @grida/daemon; the daemon is AI-free by contract (pinned by packages/grida-daemon/src/__boundary__.test.ts).

Use desktop for Electron shell work: BrowserWindow, menus, preload, native dialogs, file associations, deep links, and CDP verification.

Adjacent: sdk-design and sdk-seam for exported contracts and host seams, security when a change touches a GRIDA-SEC boundary.

When to use this skill

• Editing packages/grida-daemon/** or packages/grida-ai-agent/**.
• Changing DaemonServer lifecycle, the tenant seam, server capabilities, HTTP routes, transport, Basic Auth, Referer policy, sessions, workspaces/files, recents, secrets, BYOK providers, model tiers, sandbox policy, shell policy, runtime, tools, prompts, skills, or message streams.
• Touching the desktop sidecar protocol because the daemon contract changed.
• Debugging empty file lists, missing sessions, broken stream parts, provider resolution, tool-call behavior, or agent state recovery.
• Editing renderer agent chat transport when the bug is in stream/session semantics rather than Electron bridge presence.

Skip this skill for BrowserWindow, menu, native dialog, file association, deep-link, packaging, or preload-only bugs. Use desktop for those.

Shape

Host adapter (Desktop today, other hosts later)
  - supplies auth material, workspace roots, entitlement hooks, sandbox wrapper
  - starts the composed daemon (createAgentDaemon) as sidecar or CLI daemon
        |
        v
DaemonServer (@grida/daemon/server)
  - loopback Hono perimeter (CORS -> Referer -> Basic Auth)
  - host capability routes: files, recents, workspaces + secrets store
  - the DaemonTenant seam (static typed list, not a plugin registry)
        |
        v
Agent tenant (@grida/agent/server — createAgentTenant)
  - AI route groups: /agent, /events, /sessions, /secrets, /providers,
    /images, /video
  - sessions SQLite, endpoint configs, runtime + run loop, stream registry
        |
        v
Runtime-agnostic agent core (@grida/agent)
  - protocol DTOs, prompts, tiers, skills, toolset, AI-SDK UI-message stream

The host owns process supervision and native capability exposure. The daemon package owns the perimeter and host capabilities; the agent package owns agent-system semantics.

Package Boundaries

Two packages, each with intentional entrypoints.

@grida/daemon (the host layer — AI-free by contract):

• @grida/daemon: handshake vocabulary (DaemonCapabilities, DAEMON_PROTOCOL), local-resource DTOs.
• @grida/daemon/server: Node-only DaemonServer, buildServer, the DaemonTenant/DaemonServices seam, Daemon discovery, and the tenant toolkit (WorkspaceRegistry, workspaceFs, SecretsStore, shell runner, request validation).
• @grida/daemon/transport: Basic Auth signing, fetch/SSE helpers, and DaemonTransport.Client (daemon route groups).
• @grida/daemon/sandbox: the sandbox policy frame (buildDaemonSandboxPolicy).

@grida/agent (the agent tenant — depends on @grida/daemon):

• @grida/agent: neutral contracts, provider metadata, run/stream DTOs, createAgent, prompt composition, toolset, tiers, and session row types.
• @grida/agent/server: Node-only createAgentTenant + createAgentDaemon (the composed server hosts run).
• @grida/agent/transport: AgentTransport.Client extends DaemonTransport.Client with the tenant routes (sessions, run/stream, events, secrets, providers, images, video).
• @grida/agent/sandbox: composed policy (buildAgentDaemonSandboxPolicy — daemon frame + AI upstream hosts).
• @grida/agent/fs: storage-agnostic virtual filesystem and AI-SDK file tools.
• @grida/agent/fs/backends/opfs: browser OPFS backend.
• @grida/agent/todos: plan store and todo_write.
• @grida/agent/tiers: model tier constants.

Keep browser-safe imports neutral. Node-only code must stay behind server, sandbox, or host adapter entrypoints. Never add an AI import to @grida/daemon — the boundary test fails, and the change belongs in the tenant. A non-AI host capability (a new file route, a viewer backend) belongs in @grida/daemon, never here.

Do not create a second desktop-specific agent implementation under desktop/src/**. Desktop supervises and adapts the composed daemon; it does not own the agent semantics.

Ownership Rules

Core tests belong in the owning package first. Perimeter, files, workspaces, secrets-store, and seam bugs get tests under packages/grida-daemon/src/**/*.test.ts; session, provider, runtime, tool, and stream-formatting bugs under packages/grida-ai-agent/src/**/*.test.ts.

Desktop tests should prove Electron imports, starts, and wires the core. They should not duplicate daemon or tenant behavior.

Anti-goals to preserve:

• Not a general model-provider router.
• Not a hosted model gateway.
• Not a billing or entitlement engine.
• Not a multi-agent orchestration graph.
• Not a UI framework.
• Not a desktop bridge package.
• Not a private chat-history IR; session rows are the contract.

Common Diagnostics

Empty list_files in a workspace-bound agent usually means the workspace filesystem was not hydrated. For design-agent workspace runs, @grida/agent must create AgentFs(NodeFsBackend(root)) and call await fs.hydrate() before tool calls.

An agent that sees SOME files but is missing whole subtrees (a .canvas deck, or most of the repo) is almost always hydrate-scan truncation, not a bug in those files. The walk stops at SCAN_MAX_FILES (10k) / SCAN_MAX_DEPTH and warns [agent-fs] … hydrate scan hit a cap … truncated (sidecar stderr). The usual cause is a large workspace_root containing heavy dirs that are NOT in IGNORED_SCAN_DIRS — vendored toolchains / git submodules (e.g. emsdk) or .claude/worktrees (full repo copies). .gitignore is NOT consulted, so a submodule slips through. Fixes: scope the workspace to the real project subdir, or add the offender to IGNORED_SCAN_DIRS in @grida/daemon's workspaces/scan.ts.

A client-resolved tool call that hangs at input-available (the turn just ends with no result; the assistant never continues) is the server-authoritative model view dropping it. The runtime rebuilds the model's input from the PERSISTED messages (buildModelMessages over listVisibleMessages), NOT the client's array — and it drops any tool call without a terminal result. For a workspace-less session (the desktop file-window sidebar, which resolves fs tools in the renderer over the live editor), the result lives only on the client's next-request assistant message. persistIncomingTail must persist those terminal tool-result parts (it does, as of the file-window sidebar fix) or the call stays input-available forever and the model never sees the answer. Diagnostic tell: workspace sessions show tool_state=output-available parts in sessions.db; a no-workspace session stuck at input-available for EVERY tool call is this class, not a tool bug. (Server-resolved tools are unaffected — the recorder writes their result straight from the stream.)

Broken desktop agent calls can still be package bugs. Check whether the same operation fails through AgentTransport.Client or package tests before debugging Electron.

If a method needs secrets, keep secret reads inside the daemon process (the store is @grida/daemon's SecretsStore; the /secrets routes are tenant-registered). The renderer may check presence and set/delete BYOK keys, but it must not receive raw secret values.

If a change adds host-specific behavior, define the strict host capability contract first. Do not let Node, Electron, or renderer-only imports leak into neutral package entrypoints.