feat(agent): tinyagents 1.5 migration wave — vendored SDK, dual-write sessions, goals/todos shadows, journals, middleware dedupe

.github/workflows/release-production.yml

@@ -372,6 +372,11 @@ jobs:
         with:
           ref: ${{ needs.prepare-build.outputs.build_ref }}
           fetch-depth: 1
+      # Targeted init (not `submodules: true`) so we skip the large tauri-cef
+      # fork the core image doesn't need. The Dockerfile COPYs vendor/ because
+      # [patch.crates-io] resolves tinyagents from vendor/tinyagents.
+      - name: Init tinyagents submodule
+        run: git submodule update --init vendor/tinyagents
       - name: Set up Docker Buildx
         uses: docker/setup-buildx-action@v3
       - name: Log in to GHCR

.github/workflows/release-staging.yml

@@ -265,6 +265,11 @@ jobs:
         with:
           ref: ${{ needs.prepare-build.outputs.build_ref }}
           fetch-depth: 1
+      # Targeted init (not `submodules: true`) so we skip the large tauri-cef
+      # fork the core image doesn't need. The Dockerfile COPYs vendor/ because
+      # [patch.crates-io] resolves tinyagents from vendor/tinyagents.
+      - name: Init tinyagents submodule
+        run: git submodule update --init vendor/tinyagents
       - name: Set up Docker Buildx
         uses: docker/setup-buildx-action@v3
       - name: Build image (no push)

.gitignore

@@ -126,3 +126,4 @@ distribution.cer
 # Release note previews
 CHANGELOG.preview.md
 *.profraw
+*.diff

.gitmodules

@@ -5,3 +5,6 @@
 [submodule "app/src-tauri/vendor/tauri-plugin-notification"]
 	path = app/src-tauri/vendor/tauri-plugin-notification
 	url = https://github.com/tinyhumansai/tauri-plugin-notification.git
+[submodule "vendor/tinyagents"]
+	path = vendor/tinyagents
+	url = https://github.com/tinyhumansai/tinyagents

Cargo.toml

@@ -42,7 +42,7 @@ crate-type = ["rlib"]
 tinyplace = "1.0.1"
 # tinyflows — host-agnostic workflow engine (typed node graph → validate → compile →
 # run on tinyagents). Powers the "Workflows" feature via the seam in
-# `src/openhuman/tinyflows/` + the `flows::` domain. Pulls tinyagents 1.3 transitively
+# `src/openhuman/tinyflows/` + the `flows::` domain. Pulls tinyagents 1.5 transitively
 # (same version openhuman already uses — no conflict). Published on crates.io.
 tinyflows = "0.3"
 # TinyAgents — Rust LLM orchestration framework (LangGraph/LangChain-style):
@@ -56,7 +56,7 @@ tinyflows = "0.3"
 # aligned to 0.40, avoiding duplicate `links = "sqlite3"` native bindings.
 # Durable graph checkpoints still use `SqlRunLedgerCheckpointer` until the
 # migration re-points those rows to the crate checkpointer.
-tinyagents = { version = "1.3", features = ["sqlite"] }
+tinyagents = { version = "1.5.0", features = ["sqlite"] }
 # TokenJuice code compressor — AST-aware signature extraction. Optional (C build)
 # behind the default `tokenjuice-treesitter` feature; disabling it falls back to
 # the language-agnostic brace-depth heuristic. See src/openhuman/tokenjuice/compressors/code.rs.
@@ -230,7 +230,7 @@ fantoccini = { version = "0.22.0", optional = true, default-features = false, fe
 serde-big-array = { version = "0.5", optional = true }
 pdf-extract = "0.10"
 # WhatsApp Web — upstream `whatsapp-rust` 0.5. Its Diesel-backed sqlite-storage
-# feature links sqlite3 separately from rusqlite 0.40, so the TinyAgents 1.3
+# feature links sqlite3 separately from rusqlite 0.40, so the TinyAgents 1.5
 # baseline compiles this provider against wacore's in-memory Backend until a
 # rusqlite-backed durable store lands.
 whatsapp-rust = { version = "0.5", optional = true, default-features = false, features = ["tokio-runtime"] }
@@ -331,6 +331,12 @@ unexpected_cfgs = { level = "warn", check-cfg = ['cfg(coverage)'] }
 # See: https://github.com/tinyhumansai/openhuman/issues/273
 [patch.crates-io]
 whisper-rs-sys = { git = "https://github.com/tinyhumansai/whisper-rs-sys.git", branch = "main" }
+# TinyAgents is vendored as a git submodule (pinned at the released tag) so
+# migration work can change the SDK source in-tree, test it against OpenHuman
+# immediately, and PR the diff upstream from the submodule. Keep the submodule
+# version in lockstep with the `tinyagents` requirement above. After cloning:
+# `git submodule update --init vendor/tinyagents` (worktrees included).
+tinyagents = { path = "vendor/tinyagents" }
 
 # Emit just enough DWARF in release builds for Sentry to symbolicate Rust
 # panics + render surrounding source lines. `line-tables-only` keeps the

Dockerfile

@@ -45,6 +45,11 @@ WORKDIR /build
 
 # Cache dependencies — copy only manifests first
 COPY Cargo.toml Cargo.lock rust-toolchain.toml ./
+# Vendored TinyAgents SDK (git submodule; [patch.crates-io] points here, so
+# the dep-cache build below already resolves it). CI must init the submodule
+# before docker build — see the "Init tinyagents submodule" steps in
+# release-production.yml / release-staging.yml.
+COPY vendor/ vendor/
 # Create a dummy src to build deps
 RUN mkdir -p src && \
     echo 'fn main() {}' > src/main.rs && \

app/src-tauri/Cargo.toml

@@ -207,6 +207,10 @@ e2e-test-support = ["openhuman_core/e2e-test-support"]
 # whisper.cpp to use MSVC's static runtime (/MT), matching CEF and avoiding
 # LNK2038/LNK1169 CRT conflicts on Windows.
 whisper-rs-sys = { git = "https://github.com/tinyhumansai/whisper-rs-sys.git", branch = "main" }
+# TinyAgents vendored submodule (repo-root vendor/tinyagents, pinned at the
+# released tag) — same patch as the root Cargo world so both resolve the
+# in-tree SDK source. `git submodule update --init vendor/tinyagents` first.
+tinyagents = { path = "../../vendor/tinyagents" }
 
 # CEF support lives on the `feat/cef` branch of tauri-apps/tauri. We carry our
 # own fork at tinyhumansai/tauri-cef on `feat/cef-notification-intercept` which

docs/tinyagents-full-migration-plan/00-baseline.md

@@ -1,21 +1,21 @@
 # 00 — Baseline: crate, features, native links
 
-Current status (2026-07-02): baseline dependency alignment is complete in both
-Cargo worlds. `tinyagents 1.3.0` is resolved with the `sqlite` feature,
+Current status (2026-07-03): baseline dependency alignment is complete in both
+Cargo worlds. `tinyagents 1.5.0` is resolved with the `sqlite` feature,
 OpenHuman pins `rusqlite = "=0.40.0"`, both worlds patch through
 `vendor/rusqlite-0.40.0` and `vendor/libsqlite3-sys-0.38.0`, and the SDK-gaps
 inventory has been refreshed against the published 1.3.0 crate source.
 
 ## Steps
 
-1. **Bump `tinyagents` to `"1.3"`** (done in both Cargo worlds — root and
+1. **Bump `tinyagents` to `"1.5.0"`** (done in both Cargo worlds — root and
    `app/src-tauri/`). Known 1.1→1.2 break already handled
    (`MessageDelta::text` ctor). Note: the `openai` crate feature was removed
    after 1.2.0 (1.2.1+ features are only `sqlite`/`repl`) — we never enabled
    it, so no impact. See "1.3.0 delta" below for new API this plan uses.
 2. **Align rusqlite to 0.40** in both worlds (`Cargo.toml` root and
    `app/src-tauri/Cargo.toml`). OpenHuman pins `rusqlite = "=0.40.0"` and
-   enables `tinyagents = { version = "1.3", features = ["sqlite"] }`.
+   enables `tinyagents = { version = "1.5.0", features = ["sqlite"] }`.
    Compatibility notes:
    - `rusqlite 0.40` and `libsqlite3-sys 0.38` are consumed directly from
      crates.io. Their build scripts use the `cfg_select!` macro (stable from

docs/tinyagents-full-migration-plan/01-tooling/README.md

@@ -5,7 +5,7 @@ exposure, and output budgeting onto SDK primitives; delete the OpenHuman
 side-lookup pattern and legacy tool plumbing.
 
 Target SDK surface (available across tinyagents 1.2.x–1.3.0; current repo lock
-is 1.3.0):
+is 1.5.0):
 
 - `Tool::policy() -> ToolPolicy { side_effects, runtime, access }` —
   serializable safety metadata (read_only/writes_files/network/destructive/

docs/tinyagents-full-migration-plan/04-sessions/03-checkpointer.md

@@ -1,6 +1,6 @@
 # 04.3 — SqliteCheckpointer swap
 
-Baseline is complete: OpenHuman is on tinyagents 1.3.0 with the `sqlite`
+Baseline is complete: OpenHuman is on tinyagents 1.5.0 with the `sqlite`
 feature and the compatible `rusqlite` pin. The remaining work is the
 OpenHuman-row migration/expiry decision.
 

docs/tinyagents-full-migration-plan/10-registry.md

@@ -37,7 +37,7 @@ aliases }`, `RegistrySnapshot` (`to_dot()`),
    native/MCP/Composio/generated tools, unsafe aliases → registry
    diagnostic errors (today: duplicate handling is scattered across generated
    tools, MCP, and native registration instead of one SDK diagnostic stream).
-   TinyAgents 1.3.0 is pinned and exposes `AliasBinding`, alias diagnostics,
+   TinyAgents 1.5.0 is pinned and exposes `AliasBinding`, alias diagnostics,
    cross-kind name-reuse detection, and `ComponentKind::{Middleware,
 Checkpointer, TaskStore, Listener}`; OpenHuman still needs to project those
    SDK diagnostics into its runtime.

docs/tinyagents-full-migration-plan/C2b-todos-parity.md

@@ -0,0 +1,103 @@
+# C2b — Task board / todos onto `graph::todos` (parity note)
+
+Status: **first slice landed** (branch `feat/tinyagents-c2-todos`). Adapter-first,
+shadow-only. Legacy stays authoritative; nothing here changes product behavior.
+
+This note pairs with the CONTINUATION plan §C2 (step 3) and records what the
+crate `tinyagents::graph::todos` surface maps onto in OpenHuman, and what it does
+**not** — the residue that must stay in the product host after a future cutover.
+
+## What landed in this slice
+
+- `src/openhuman/todos/graph_shadow.rs` — the adapter:
+  - Total, lossless status mapping OpenHuman ↔ crate (`map_status_to_crate` /
+    `map_status_from_crate`) — the two `TaskCardStatus` enums share the same
+    seven variants (`Todo`, `AwaitingApproval`, `Ready`, `InProgress`,
+    `Blocked`, `Done`, `Rejected`).
+  - `TaskBoardCard` field-by-field conversion (`to_crate_card`) — all metadata
+    (objective, plan, assignedAgent, allowedTools, approvalMode,
+    acceptanceCriteria, evidence, notes, blocker, sessionThreadId,
+    sourceMetadata, order, updatedAt) preserved.
+  - `spawn_mirror` — after every authoritative `todos::ops::save_cards` write of
+    a `Thread` board, mirrors the persisted cards into a crate `FileStore`
+    (`<workspace>/tinyagents_graph_store`) under namespace `graph.todos`. Fire-
+    and-forget, log-only; a crate rejection (e.g. the single-`InProgress`
+    invariant) is warn-logged as a **DIVERGENCE**.
+  - `spawn_shadow_claim` — wired into `todos::ops::claim_card` (which is the
+    single claim entry-point the dispatcher's two claim sites in
+    `task_dispatcher/dispatch.rs` funnel through, plus RPC/reclaim callers). It
+    seeds the crate board with the pre-claim snapshot and replays the crate
+    `claim_card` CAS, warn-logging when the crate ok/err verdict disagrees with
+    the authoritative legacy claim.
+
+The legacy claim/save path is byte-for-byte preserved: `claim_card` was
+refactored to compute one ok/err verdict via an extracted `apply_claim` helper
+(so the shadow sees the same not-found / wrong-status / invariant outcomes), but
+the persisted result and all existing tests are unchanged.
+
+## Crate `TodoTool` vs `agent/tools/todo.rs` — what maps
+
+The crate ships a single multiplexer `TodoTool` (`op`-dispatched) that is a near
+drop-in for the OpenHuman `todo` tool. Shared surface:
+
+| Concern | Crate `TodoTool` | OpenHuman `tools/todo.rs` | Match? |
+| --- | --- | --- | --- |
+| Dispatch style | single tool, `op` field | single tool, `op` field | yes |
+| Thread binding | `ToolExecutionContext::thread_id` (never an arg) | `thread_context::current_thread_id()` + `fork_context` parent | yes (both bind to current thread, never an arg) |
+| `add`/`edit`/`update_status`/`remove`/`replace`/`clear`/`list` | present | present | yes |
+| Optional card fields | objective/plan/assignedAgent/allowedTools/approvalMode/acceptanceCriteria/evidence/notes/blocker | same | yes |
+| Return shape | `{ threadId, cards, markdown }` | `{ threadId, cards, markdown }` | yes |
+| Status aliases | `parse_status` (pending→todo, approved→ready, …) | `ops::parse_status` (identical alias table) | yes |
+| Single-`InProgress` invariant | `enforce_single_in_progress` (hard error) | `enforce_single_in_progress` (identical) | yes |
+| `claim_card` CAS | `store::claim_card(expected, target)` | `ops::claim_card(expected, target)` | yes (identical semantics; proven by shadow tests) |
+
+## What does NOT map (product residue — must stay in the host)
+
+1. **Approval-gate coupling.** OpenHuman's `todo` tool stamps a default
+   `approvalMode` by reading `config.autonomy.require_task_plan_approval`
+   (`default_task_approval_mode`), and the dispatcher's
+   `requires_plan_approval` + `TaskPlanAwaitingApproval` `DomainEvent` drive the
+   interactive plan-review gate. The crate `TodoTool` has **no** config read and
+   **no** approval-gate wiring — it exposes `decide_plan`/`revise_plan` state
+   transitions only. The gate policy stays product.
+2. **`DomainEvent` emissions.** `ops::claim_card`/mutations emit
+   `AgentProgress::TaskBoardUpdated` (via `fork_context` `on_progress`) and the
+   dispatcher publishes `TaskPlanAwaitingApproval`. The crate store emits
+   nothing. All event vocabulary stays product (ledger: keep).
+3. **RPC projection shapes.** `threads.task_board_*` and `openhuman.todos_*`
+   (see `todos/schemas.rs`) are the wire contracts the kanban UI binds to
+   (`app/src/services/api/todosApi.ts`, `USER_TASKS_THREAD_ID = "user-tasks"`).
+   These are **unchanged** by this slice and, per §C2, become read-side
+   projections over the crate store only at cutover — not now.
+4. **Scratch board.** OpenHuman has a thread-less in-memory `BoardLocation::Scratch`
+   fallback (tool calls outside a chat thread). The crate board is always
+   `(Store, thread_id)`, so scratch mutations have **no** crate mirror target —
+   the shadow skips them (trace-logged).
+5. **Persistence substrate + timestamps.** Product persists RFC3339
+   `updated_at` to `<workspace>/agent_task_boards/<hex(thread_id)>.json`; the
+   crate uses epoch-millis strings in a `Store` namespace. The mirror does not
+   reconcile timestamps (cosmetic). Card-id minting also differs
+   (`task-<uuid>` product vs `task-<seq>` crate) but ids are passed through, so a
+   persisted board round-trips.
+6. **Run lifecycle.** `todos/runs.rs` (run records, heartbeats, stale-reclaim)
+   and `task_dispatcher/` executor mechanics (executor resolution, autonomous
+   run, board write-back) are **not** part of the crate todos surface — they
+   stay product and are the §C2 step-3 "runner node" work, tracked separately.
+
+## Single-writer constraint
+
+The crate `Store` has no compare-and-set, so ns `graph.todos` assumes a single
+writer. The core process is that single writer (both the mirror and the
+shadow-claim run in-core). Documented in the module header; honoured because all
+mutations funnel through `todos::ops`.
+
+## Next (not in this slice)
+
+- Flip the mirror from shadow to authoritative (crate store becomes the source
+  of truth; legacy JSON becomes a projection or is retired).
+- Reimplement `threads.task_board_*` / `openhuman.todos_*` as projections over
+  the crate store.
+- Replace the dispatcher claim/poll loop with the crate `claim_card` CAS + a
+  graph runner node, keeping `DomainEvent` emission + channel bindings product.
+- Delete `task_board.rs` + todo CRUD mechanics + dispatcher executor mechanics
+  (~3.2k + tests) once parity logs are clean.