Files
openhuman/docs/tinyagents-migration-spec.md
T

837 lines
45 KiB
Markdown

# TinyAgents Migration Spec
Status: draft migration backlog
TinyAgents source reviewed: `tinyhumansai/tinyagents` `origin/main` at
`8f226f1`, crate version `1.1.0`. Refreshed against `tinyhumansai/tinyagents`
`main` at `348a0e7dc71a1f9039f3d523a2a384661a7a9acd` after the SDK/docs update.
Current OpenHuman dependency in this checkout is
`tinyagents = { version = "1.5.0", features = ["sqlite"] }`.
OpenHuman already depends on TinyAgents and already routes the live agent turn
through `src/openhuman/tinyagents/`. This spec is not a proposal to add
TinyAgents. It is a todo list for moving the rest of OpenHuman's generic agent
runtime behavior onto TinyAgents primitives while keeping OpenHuman-owned
product semantics in OpenHuman.
Current inventory snapshot: [`tinyagents-harness-migration-audit.md`](tinyagents-harness-migration-audit.md).
## Goal
Use TinyAgents as the generic runtime for:
- model/provider abstraction and model selection
- tools and tool schemas
- middleware around model/tool calls
- streaming, events, traces, and replayable run status
- session transcript storage and migration targets
- prompt/response cache layout protection
- token usage and cost rollups
- state graphs, fanout, reducers, checkpoints, and interrupts
- sub-agent recursion, steering, cancellation, and reusable sessions
- deterministic testkit coverage for the runtime seams
OpenHuman should continue to own:
- desktop product UX and Tauri/RPC boundaries
- user/workspace config, credentials, keychain, and approval records
- OpenHuman memory stores, thread transcripts, run ledgers, and controllers
- security policy, sandboxing, tool permission tiers, and workspace roots
- product-specific built-in agents, prompts, MCP setup, Composio, channels, and
native tools
- compatibility with existing JSON-RPC method names and persisted state
## Sources Reviewed
TinyAgents SDK:
- `src/lib.rs`
- `src/harness/*`
- `src/graph/*`
- `src/registry/*`
- `src/language/*`
- `src/repl/*`
- `docs/modules/harness/*.md`
- `docs/modules/graph/*.md`
- `docs/modules/registry/*.md`
- `docs/modules/expressive-language/README.md`
- `docs/modules/repl-language/README.md`
- examples: `agent_loop_tools`, `orchestrator_subagents`, `durable_graph`,
`openai_graph_agent`, `openai_self_blueprint`
OpenHuman Rust core:
- `src/openhuman/tinyagents/*`
- `src/openhuman/agent/**`
- `src/openhuman/agent_orchestration/**`
- `src/openhuman/agent_registry/**`
- `src/openhuman/tools/**`
- `src/openhuman/inference/**`
- `src/openhuman/cost/**`
- `src/openhuman/context/**`
- `src/openhuman/approval/**`
- `src/openhuman/security/**`
- `src/openhuman/mcp_registry/**`
- `src/core/event_bus/**`
- `src/core/all.rs`
- `gitbooks/developing/architecture/agent-harness.md`
## Current Adoption Inventory
Already done or partially done:
- `Cargo.toml` pins `tinyagents = { version = "1.5.0", features = ["sqlite"] }`.
- `src/openhuman/tinyagents/mod.rs` registers OpenHuman `Provider` and `Tool`
adapters on `tinyagents::harness::runtime::AgentHarness`.
- `ProviderModel` maps OpenHuman `ChatRequest`/`ChatResponse` into
`tinyagents::harness::model::{ModelRequest, ModelResponse, ModelStream}`.
- `ToolAdapter` and `SharedToolAdapter` map OpenHuman tools into
`tinyagents::harness::tool::Tool`.
- `OpenhumanEventBridge` maps TinyAgents `AgentEvent` into `AgentProgress` and
the global cost tracker.
- `StopHookMiddleware`, `ContextCompressionMiddleware`, and
`MessageTrimMiddleware` are already used on the TinyAgents path.
- `SqlRunLedgerCheckpointer` implements TinyAgents `Checkpointer` on top of the
OpenHuman session DB while the migration re-points checkpoint rows to the
crate checkpointer.
- `spawn_parallel_graph` uses `GraphBuilder` and
`graph::parallel::map_reduce` for reusable concurrent fanout.
- `model_council`, `workflow_runs`, `agent_teams`, and
`tinyagents/delegation.rs` already use TinyAgents graphs.
- Built-in agents already have `graph.rs` selectors, but most return
`AgentGraph::Default`.
Important current gaps:
- OpenHuman still has separate registries for agents, tools, MCP tools, model
providers, controllers, cost, and event bus projections.
- Tool safety metadata exists in OpenHuman traits but is not fully expressed as
TinyAgents tool safety/runtime metadata.
- Cost/usage is still converted through a bridge, not an end-to-end TinyAgents
usage/cost journal.
- Event streams are mirrored into `AgentProgress`, but TinyAgents event journals
and status stores are not the canonical durable inspection surface.
- Provider model profiles, model resolution, and fallback remain primarily in
OpenHuman inference/router logic.
- Sub-agent lifecycle, durable state, worker threads, and wait/abort controls
still live in OpenHuman orchestration stores.
Some todos below are local adapter work. Others still require upstream
TinyAgents SDK extensions. In particular, the SDK has a strong tool/runtime
boundary today (`ToolSchema`, `ToolExecutionContext`, middleware hooks), but
OpenHuman's full tool safety metadata is richer than the current SDK schema.
After the TinyAgents `main` refresh, durable session/cache/journal primitives are
no longer the broad SDK gap they were in the original baseline; the remaining
work is to design OpenHuman's compatibility adapter, migrate old transcripts and
run-ledger rows, and prove restart/resume parity.
## Migration Rules
- Keep every product-facing JSON-RPC contract stable unless a migration plan is
written next to the code change.
- Do not bypass OpenHuman approval, security policy, sandbox, workspace root, or
credential boundaries by adopting a generic TinyAgents tool API.
- Prefer adapters first, then flip ownership once tests prove parity.
- Preserve existing transcript and run-ledger compatibility. TinyAgents may
become the internal runtime without changing persisted public records in the
same PR.
- Old OpenHuman session JSONL/Markdown data should move through a one-time,
idempotent migration script into TinyAgents-compatible store/journal records
before the old readers are deleted.
- Every migration task needs unit coverage plus at least one JSON-RPC or
harness-level e2e when behavior crosses controller, tool, provider, or graph
boundaries.
## Phase 0 - Baseline And Drift Control
- [x] Add a version/feature compatibility note to the OpenHuman architecture doc.
- OpenHuman files: `gitbooks/developing/architecture/agent-harness.md`,
`Cargo.toml`.
- TinyAgents components: crate features `default`, `openai`, `sqlite`, `repl`.
- Acceptance: document why default features are used, why TinyAgents `sqlite`
is enabled through the aligned `rusqlite` stack, and which OpenHuman adapters
still replace SDK-owned providers.
- **Done:** added "TinyAgents crate: features & compatibility" section to
agent-harness.md (default-only, `openai`/`sqlite`/`repl` rationale, adapter
map) + fixed stale `council_graph.rs`/`member_graph.rs` links.
## Phase 1 - Tools
- [~] Make OpenHuman tool metadata round-trip into TinyAgents tool metadata.
- OpenHuman files: `src/openhuman/tools/traits.rs`,
`src/openhuman/tinyagents/tools.rs`, `src/openhuman/tinyagents/convert.rs`.
- TinyAgents components: `harness::tool::{ToolSchema, ToolFormat,
ToolExecutionContext, ToolResult}`.
- Migrate: permission level, external effect, generated runtime context,
timeout policy, concurrency safety, result-size cap, display label/detail,
markdown support.
- Acceptance: a TinyAgents tool call has enough metadata for middleware to
enforce approval, security, timeout, concurrency, truncation, and display
behavior without re-querying OpenHuman trait methods ad hoc.
- **1.3 update:** crate `ToolPolicy`, `Tool::policy()`,
`ToolRegistry::policies()`, and `ToolPolicyMiddleware` now provide the
SDK-owned safety/runtime/access projection. `ToolSchema` still carries only
name/description/parameters/format — it has **no** model-visible
metadata/extension map — so OpenHuman should map enforcement fields into
`ToolPolicy` while keeping display/schema annotations as app-side metadata.
Existing side-lookup middleware (`ApprovalSecurityMiddleware`,
`ToolOutputMiddleware`) can shrink as those policy snapshots become the
source of truth.
- [x] Move unknown-tool recovery into a reusable middleware or tool policy layer.
- Current path: `run_policy_for` sets
`UnknownToolPolicy::ReturnToolError`; no sentinel tool is registered.
- TinyAgents components: `ToolRegistry`, `ToolMiddleware`,
`AgentEvent::ToolStarted/ToolCompleted`, repairable tool results.
- Acceptance: hallucinated tool names remain recoverable, sub-agent wording is
preserved, and TinyAgents event stream records the original requested tool
name without exposing the sentinel as a model-visible tool.
- **1.3 update:** crate `RunPolicy.unknown_tool` now has
`UnknownToolPolicy::{Fail, ReturnToolError, Rewrite}` and emits
`AgentEvent::UnknownToolCall` with the original requested name/arguments.
OpenHuman now uses that policy directly; `UNKNOWN_TOOL_SENTINEL` and
`UnknownToolRewriteMiddleware` are gone from source.
- [x] Route approval and security through TinyAgents middleware.
- Current OpenHuman files: `src/openhuman/approval/*`,
`src/openhuman/security/*`, `src/openhuman/tinyagents/tools.rs`.
- TinyAgents components: `ToolMiddleware`, `ToolExecutionContext`,
tool safety metadata.
- Acceptance: approval checks happen in `before_tool`/`wrap_tool`, emit typed
events, preserve audit rows, and return model-consumable denial results.
- **Done:** `ApprovalSecurityMiddleware` (`tinyagents/middleware.rs`, a
`wrap_tool` middleware) replaces the inline approval block in
`execute_openhuman_tool`. Denials short-circuit with a model-consumable
result; approved external-effect calls now record a terminal audit row
(`record_execution`) the old path dropped. Typed approval events still ride
`DomainEvent` (the crate `AgentEvent` enum has no approval variant — SDK gap).
Tool-*internal* security (path/command `live_policy`) stays per-tool by
design. Follow-ups: channel permission-ceiling threading; per-tool metadata
side-lookup (Task C).
- [ ] Use TinyAgents bounded-concurrent tool execution where safe.
- OpenHuman files: `src/openhuman/tools/traits.rs`,
`src/openhuman/tinyagents/tools.rs`.
- TinyAgents components: graph `Send`, graph fanout, or harness tool
execution policy.
- Acceptance: read-only/concurrency-safe tool batches can run in parallel
with deterministic result ordering and identical transcript semantics.
## Phase 2 - Models And Providers
- [ ] Register OpenHuman inference providers as TinyAgents model registry entries.
- OpenHuman files: `src/openhuman/inference/provider/*`,
`src/openhuman/inference/model_ids.rs`, `src/openhuman/tinyagents/model.rs`.
- TinyAgents components: `harness::model::{ChatModel, ModelRegistry,
ModelProfile, CapabilitySet, ModelRequest, ModelResponse}`.
- Acceptance: every workload route (`agentic`, `reasoning`, `coding`,
`memory`, `subconscious`, etc.) can resolve to a TinyAgents model entry
while retaining OpenHuman provider strings and config compatibility.
- [~] Translate OpenHuman provider capability data into TinyAgents model profiles.
- OpenHuman files: `src/openhuman/inference/provider/traits.rs`,
`src/openhuman/inference/provider/factory.rs`,
`docs/inference-provider-catalog.md`.
- TinyAgents components: `ModelProfile`, `CapabilitySet`, registry model
catalog.
- Acceptance: context window, tool calling, streaming, vision, structured
output, reasoning, local/cloud source, and provider-family metadata are
available before dispatch.
- **Partial:** every `ProviderModel` registered by the shared runner now
carries a crate `ModelProfile` built at construction from the provider's
canonical capability accessors — tool calling (+parallel), vision
(`modalities.image_in`), streaming, local/remote source — plus the
runner-threaded token limits (`with_context_window` → `max_input_tokens`,
output cap → `max_output_tokens`). `ChatModel::profile()` returns it, so
the crate's pre-dispatch validation and structured-output strategy see real
capabilities. Remaining: structured-output/JSON-schema/reasoning flags
(no OpenHuman capability source yet), release/status metadata, and a
registry-level model *catalog* (ties into the workload-route registry item
above).
- [ ] Move model fallback and retry policy to TinyAgents policy/middleware.
- OpenHuman files: `src/openhuman/inference/provider/reliable.rs`,
`src/openhuman/inference/provider/router.rs`,
`src/openhuman/tinyagents/mod.rs`.
- TinyAgents components: `RunPolicy`, `RetryPolicy`, `FallbackPolicy`,
`ModelFallbackMiddleware`, `AgentEvent::RetryScheduled`,
`AgentEvent::FallbackSelected`.
- Acceptance: OpenHuman provider retry does not double-retry under
TinyAgents, fallback events are typed, and tests cover transient 429/5xx,
config rejection, and billing exhaustion.
- [ ] Preserve provider-specific metadata in the TinyAgents message model.
- OpenHuman files: `src/openhuman/inference/provider/traits.rs`,
`src/openhuman/tinyagents/convert.rs`,
`src/openhuman/tinyagents/model.rs`.
- TinyAgents components: `ContentBlock`, `AssistantMessage`, provider
metadata, tool-call ids.
- Acceptance: Gemini thought signatures, reasoning content, native tool-call
ids, cached tokens, and raw provider metadata survive multi-turn history.
## Phase 3 - Middleware
- [ ] Convert OpenHuman turn cross-cuts into named TinyAgents middleware.
- Current OpenHuman surfaces: stop hooks, approval gate, security policy,
output caps, context compression, memory injection, tool allowlists,
cost/usage, prompt cache stability, event bridge.
- TinyAgents components: `Middleware`, `ModelMiddleware`, `ToolMiddleware`,
`MiddlewareStack`, `RunContext`.
- Acceptance: each cross-cut has a stable middleware name, tests for ordering,
emitted events, and explicit interaction with streaming/retry/fallback.
- [ ] Add OpenHuman policy middleware for dynamic tool exposure.
- OpenHuman files: `src/openhuman/agent_registry/*`,
`src/openhuman/agent/harness/subagent_runner/**`,
`src/openhuman/tools/user_filter.rs`.
- TinyAgents components: `before_model`, `before_tool`, `ToolRegistry`.
- Acceptance: agent `tool_allowlist`, `tool_denylist`, sub-agent tool scope,
MCP tool visibility, and channel permission ceilings are enforced through
middleware rather than scattered pre-filtering.
- [ ] Add prompt/cache-layout middleware tests.
- OpenHuman files: `src/openhuman/context/*`,
`src/openhuman/agent/harness/session/turn/core.rs`,
`src/openhuman/tinyagents/summarize.rs`.
- TinyAgents components: `harness::cache::{CachePolicy, CacheLayoutEvent,
PromptCacheLayout, ResponseCache}`, `ContextCompressionMiddleware`,
`MessageTrimMiddleware`.
- Acceptance: system prompt prefix remains stable across later turns; volatile
memory, timestamps, tool results, and steering messages land in the tail.
- [ ] Move OpenHuman response-cache and provider KV-cache protection onto
TinyAgents cache primitives.
- OpenHuman files: `src/openhuman/agent/harness/session/turn/core.rs`,
`src/openhuman/context/*`, `src/openhuman/tinyagents/middleware.rs`.
- TinyAgents components: `harness::cache::{ResponseCache,
PromptCacheLayout, CachePolicy}`, `AgentEvent::CacheHit`,
`AgentEvent::CacheMiss`.
- Acceptance: repeated deterministic model requests can be served by the
TinyAgents response cache where safe, prompt-prefix stability is asserted by
`PromptCacheLayout`, and OpenHuman cache-align warnings become TinyAgents
cache-layout events.
## Phase 4 - Events, Status, And Observability
- [ ] Make TinyAgents event journals the canonical internal run event stream.
- OpenHuman files: `src/openhuman/tinyagents/observability.rs`,
`src/core/event_bus/*`, `src/openhuman/notifications/*`,
`src/openhuman/session_db/run_ledger/*`.
- TinyAgents components: `HarnessEventJournal`, `HarnessStatusStore`,
`GraphEventJournal`, `GraphStatusStore`, `harness::store::AppendStore`,
`harness::store::JsonlAppendStore`, `AgentEvent`, `GraphEvent`.
- Acceptance: UIs can reconstruct a running or completed agent turn from
persisted TinyAgents events without relying only on transient
`AgentProgress`.
- [x] Write a one-time OpenHuman session transcript migration into TinyAgents
store/journal records. Done: `src/openhuman/session_import/`
(`openhuman.session_import_run`), design + as-built notes in
`docs/tinyagents-session-migration-design.md`. Write-only Phase 1; read-side
shadow/cutover remain.
- OpenHuman files: `src/openhuman/agent/harness/session/transcript.rs`,
`src/openhuman/agent/harness/session/migration.rs`, user workspace
`session_raw/*.jsonl`, legacy Markdown session directories.
- TinyAgents components: `harness::store::{Store, AppendStore, FileStore,
JsonlAppendStore}`, `harness::message::Message`, harness event/status
records with `thread_id`, `run_id`, `root_run_id`, and stream offsets.
- Acceptance: old OpenHuman sessions are imported idempotently, preserving
timestamps, transcript stems, parent/child session links, provider/model
metadata, tool-call ids, native/XML/P-format history, and malformed-file
warnings. Existing OpenHuman readers remain as compatibility projections
until parity fixtures prove the migration.
- [ ] Bridge TinyAgents events into `DomainEvent` as a compatibility projection.
- OpenHuman files: `src/core/event_bus/events.rs`,
`src/openhuman/agent/bus.rs`, `src/openhuman/tinyagents/observability.rs`.
- Acceptance: existing subscribers continue to receive `DomainEvent`, but new
code reads TinyAgents events/status first.
- [ ] Persist graph run status and checkpoint metadata in OpenHuman run ledger.
- OpenHuman files: `src/openhuman/tinyagents/checkpoint.rs`,
`src/openhuman/session_db/run_ledger/store.rs`,
`src/openhuman/agent_orchestration/**`.
- TinyAgents components: `Checkpoint`, `CheckpointMetadata`,
`GraphRunStatus`, `GraphObservation`.
- Acceptance: command center, workflow runs, delegation, and team runs can
list checkpoints, current node/task status, and replay offsets from one DB
source.
- [x] Export graph topology for debugging and UI inspection.
- OpenHuman files: built-in `graph.rs` files, `agent_orchestration/*/graph.rs`,
`model_council/graph.rs`.
- TinyAgents components: `GraphTopology`, `to_json`, `to_mermaid`,
validation report.
- Acceptance: every custom OpenHuman graph has a debug endpoint or test
snapshot that exports topology and validates missing nodes/routes.
- **Done:** `tinyagents/topology.rs` — `GraphTopologyReport` (mermaid + JSON +
validation errors/warnings), `describe()`, and `all_graph_topologies()`.
Pattern: each graph exposes a `build_*_graph` (structure) reused by both the
runner and a `*_topology()` that builds it with no-op stub closures and
returns `CompiledGraph::topology()`. Exported graphs: `agent_teams:member`,
`delegation` (extracted `build_delegation_graph`),
`workflow_runs:scheduler` (`build_scheduler_graph` with injected
`select`/`run` engine effects), `subagent:pipeline`, and
`spawn_parallel_agents`. Generic map-reduce fan-outs such as council runs
are still item-count-driven dispatch→N→collect patterns, not fixed named
topologies, so they are intentionally not exported. Debug endpoint:
`agent.graph_topologies` JSON-RPC controller (`agent/schemas.rs`) returning
`{name, ok, errors, warnings, mermaid, topology}` per graph.
## Phase 5 - Usage, Cost, And Budgets
- [~] Replace bridge-only usage accounting with TinyAgents usage records.
- OpenHuman files: `src/openhuman/cost/*`,
`src/openhuman/tinyagents/observability.rs`,
`src/openhuman/inference/provider/traits.rs`.
- TinyAgents components: `harness::usage::{Usage, UsageTotals}`,
`harness::cost::CostTotals`, `AgentEvent::UsageRecorded`.
- Acceptance: input, output, cached input, reasoning, image/audio, embedding,
tool/model call counts, and estimated/provider-reported source are recorded
in normalized records.
- **Partial (real bug fixed):** the bridge hardcoded `charged_amount_usd: 0.0`
and `ProviderModel` dropped cached tokens, so EVERY tinyagents turn recorded
**$0 cost**. Now `model.rs` carries `cached_input_tokens` via crate
`Usage.cache_read_tokens`, and the bridge estimates per-call cost from
catalogued per-MTok rates (`cost::catalog::estimate_cost_usd`). Remaining:
reasoning/image/audio/embedding token fields, model/tool call counts, and an
explicit estimated-vs-provider-charged `cost_source` tag on `TokenUsage`
(provider-charged preservation needs an out-of-band carry — crate `Usage` has
no USD field).
- [~] Move budget checks to pre-call TinyAgents middleware.
- OpenHuman files: `src/openhuman/cost/tracker.rs`,
`src/openhuman/tinyagents/mod.rs`.
- TinyAgents components: `RunPolicy`, cost middleware, `before_model`,
`before_tool`.
- Acceptance: per-run, per-thread, daily, and monthly budgets can warn or
fail before spend where enough data exists, then reconcile after provider
usage is known.
- **Partial:** `CostBudgetMiddleware` (`before_model`) fails the run before a
model call when the global daily/monthly budget is already exceeded
(`CostTracker::check_budget`), and logs on the warning threshold. Self-gating
on `config.enabled`; previously daily/monthly enforcement was dormant on the
tinyagents path. Remaining: per-run/per-thread budgets (need new
`CostConfig` fields + thread-id threading into the runner) and projecting the
*next* call's cost pre-spend (needs an input-token estimate).
- [~] Add cost rollup across sub-agents and graphs.
- OpenHuman files: `src/openhuman/agent_orchestration/**`,
`src/openhuman/cost/global.rs`.
- TinyAgents components: run ids, parent/root run lineage,
`SubAgentStarted`, `SubAgentCompleted`, graph child runs.
- Acceptance: parent run totals include child agent/model/tool usage without
double counting dashboard totals.
- **Partial (audit + real gap fixed).** Audit of the current mechanics:
(1) parent-turn rollup — the `turn_subagent_usage` task-local collector
wraps the turn future; `run_typed_mode` (the single sub-agent chokepoint)
records every inline child, and graph fan-outs (`spawn_parallel_graph`,
delegation, council) execute via `join_all` **on the same task**, so their
children inherit the collector too; (2) dashboard totals — the global
tracker is fed per model call by each run's own event bridge, and the
parent's fold into `LastTurnUsage`/transcript never re-records to the
tracker, so there is no double counting. **Gap fixed:** the tracker feed
lived *only* in the bridge, so an unobserved (`on_progress = None`,
fire-and-forget) turn's spend never reached the dashboard — the runner's
cost fallback now records the aggregate via `record_unobserved_turn_usage`
(mutually exclusive with the bridge → exactly-once). Remaining: crate
run-id / parent-root lineage on cost records (needs `TokenUsage` schema
fields), and rollup for *detached* background children beyond
global-tracker capture (documented behavior today).
## Phase 6 - Graph Runtime And Orchestration
- [ ] Convert remaining ad hoc control loops into explicit TinyAgents graphs.
- Candidate OpenHuman files: `src/openhuman/agent_orchestration/*`,
`src/openhuman/subconscious/*`, `src/openhuman/cron/*`,
`src/openhuman/learning/*`, `src/openhuman/tools/ops.rs`.
- TinyAgents components: `GraphBuilder`, `Command`, conditional routing,
reducers, `Send`, barriers, recursion policy.
- Acceptance: every long-running multi-step orchestration has named nodes,
route tests, recursion bounds, cancellation checks, and topology export.
- [ ] Replace simple fanout helpers with graph `Send` where payload-specific
fanout matters.
- Current helper: `src/openhuman/tinyagents/orchestration.rs`.
- TinyAgents components: `Command::send`, `GraphInput`, `NodeContext::send_arg`.
- Acceptance: map-reduce style flows can schedule multiple invocations of the
same node with distinct payloads instead of materializing one node per item.
- [ ] Make `spawn_parallel_agents` a first-class TinyAgents graph tool.
- Current OpenHuman files:
`src/openhuman/agent_orchestration/tools/spawn_parallel_agents.rs`,
`src/openhuman/tinyagents/orchestration.rs`,
`src/openhuman/agent_orchestration/worktree.rs`,
`src/openhuman/agent_orchestration/workflow_runs/engine.rs`.
- Current behavior: validates at least two tasks, checks parent context,
enforces `max_parallel_tools`, resolves `AgentDefinition`s, enforces the
parent `subagents.allowlist`, optionally creates per-worker git worktrees,
fans workers out through `spawn_parallel_graph` + `map_reduce`, collects results in input
order, emits `DomainEvent` + `AgentProgress`, detects stale parent file
reads and cross-worker changed-file overlaps, and returns a structured
`parallel_agents` JSON payload.
- TinyAgents components: `GraphBuilder`, `Command`, `Send`,
`NodeContext::send_arg`, `ChannelSet`/reducers, `GraphEventSink`,
`SubAgentNode` or OpenHuman `run_subagent` adapter, `RecursionPolicy`,
`RunPolicy`, `CancellationToken`.
- Required migration shape:
- `validate` node: parse tasks, enforce min/max count, parent context,
allowlist, toolkit requirements, and worktree preflight.
- `dispatch` node: use `Send` to schedule one worker invocation per task
instead of generating one static worker node per task.
- `worker` node: run the OpenHuman sub-agent build pipeline with inherited
model/tool/security policy, optional `worktree_action_dir`, child task id,
and bounded turn/output budgets.
- `collect` reducer: aggregate successes, failures, elapsed time,
iterations, worktree status, changed files, and stale read markers in
deterministic task order.
- `finalize` node: emit compatibility `DomainEvent`/`AgentProgress`
projections, overlap warnings, and the existing JSON result shape.
- Acceptance: parallel agent runs are checkpointable, cancellable at graph
boundaries, bounded by parent policy, observable through TinyAgents graph
events/status, compatible with existing `spawn_parallel_agents` tool
output, and able to run edit-capable workers in isolated worktrees without
silently falling back to shared workspace.
- [ ] Define parallel-agent ownership and scheduling policy explicitly.
- OpenHuman files: `src/openhuman/agent_registry/agents/orchestrator/prompt.md`,
`src/openhuman/agent_orchestration/tools/spawn_parallel_agents.rs`,
`src/openhuman/tools/traits.rs`.
- TinyAgents components: graph route metadata, `RunPolicy`, task metadata,
tool safety metadata.
- Required policy:
- The parent must provide disjoint ownership boundaries for write-capable
tasks, or the graph rejects/falls back to serial delegation.
- Read-only workers may share the parent workspace; write-capable workers
should request `isolation = "worktree"`.
- Parent and children inherit one root run id for cost/event rollups, but
each child gets its own task id and optional worker-thread id.
- A child cannot widen tools, model choice, sandbox mode, trusted roots, or
budget beyond the parent-granted policy.
- Cancellation, steering, and wait/collect must be delivered at graph or
harness safe boundaries only.
- Acceptance: the orchestrator can ask for parallelism without prompt-only
conventions; policy violations become structured graph/tool errors.
- [ ] Make per-agent `graph.rs` selectors real customization points.
- OpenHuman files: `src/openhuman/agent_registry/agents/*/graph.rs`,
`src/openhuman/agent/harness/agent_graph.rs`.
- TinyAgents components: `CompiledGraph`, sub-agent nodes, graph testkit.
- Acceptance: at least three agents get bespoke graphs where useful:
orchestrator, researcher, and tool_maker are good first candidates.
- [~] Keep durable orchestration stores OpenHuman-owned until the OpenHuman
compatibility adapter is written.
- Current OpenHuman files: `running_subagents.rs`, `workflow_runs`,
`agent_teams`, `command_center`, `subagent_sessions`.
- Updated TinyAgents status: current `main` has harness stores, JSONL append
journals, lineage-aware harness/graph status, `SubAgentSession`, subgraph
nodes, and sub-agent graph nodes. This means the blocker is no longer only
"SDK lacks storage"; it is now the OpenHuman adapter/migration design and
restart/resume parity.
- Acceptance: migrate durable SQL/JSON state through a compatibility adapter,
not by dropping records into in-memory task storage. OpenHuman controllers
continue to read compatible projections while TinyAgents records become the
canonical internal state.
- [ ] Add graph interrupt/resume for human review points.
- OpenHuman files: `src/openhuman/approval/*`,
`src/openhuman/agent_orchestration/workflow_runs/*`,
`src/openhuman/tinyagents/delegation.rs`.
- TinyAgents components: `Interrupt`, `ResumeTarget`, `Command::resume`,
checkpoints.
- Acceptance: approval/review pauses are durable graph interrupts where the
run can resume from the exact checkpoint after user action.
## Phase 7 - Sub-Agents, Steering, And Recursion
- [ ] Represent `spawn_subagent`, `steer_subagent`, `wait_subagent`, and
follow-ups as TinyAgents steering commands plus OpenHuman projections.
- OpenHuman files: `src/openhuman/agent_orchestration/tools.rs`,
`src/openhuman/agent_orchestration/running_subagents.rs`,
`src/openhuman/agent/harness/subagent_runner/**`.
- TinyAgents components: `SteeringCommand`, `SteeringHandle`,
`SubAgentSession`, `SubAgentTool`, recursion depth events.
- Acceptance: mid-flight messages are delivered only at safe loop boundaries,
accepted/rejected steering emits events, and tool/model allowlists can only
narrow from parent policy.
- [ ] Re-express the OpenHuman sub-agent pipeline as a TinyAgents subgraph.
- OpenHuman files: `src/openhuman/agent/harness/subagent_runner/**`,
`src/openhuman/agent_orchestration/tools/spawn_subagent.rs`,
`src/openhuman/agent_orchestration/subagent_sessions/**`.
- TinyAgents components: `harness::subagent::{SubAgent, SubAgentSession,
SubAgentTool}`, `graph::subagent_node::{SubAgentNode, SubAgentPolicy}`,
`graph::subgraph::{shared_subgraph_node, adapter_subgraph_node}`,
`CapabilityRegistry`, graph status/observability.
- Acceptance: definition resolution, tool filtering, prompt assembly,
toolkit preflight, sandbox/action-root narrowing, handoff cache,
checkpoint/awaiting-user handback, and worker-thread mirroring are explicit
graph nodes or adapters. The final child run has TinyAgents lineage, status,
usage/cost rollup, and transcript storage; OpenHuman keeps only product
policy nodes and compatibility response formatting.
- [ ] Reconcile OpenHuman spawn depth with TinyAgents recursion policy.
- OpenHuman files: `src/openhuman/agent/harness/spawn_depth_context.rs`,
`src/openhuman/agent/harness/subagent_runner/**`.
- TinyAgents components: `RecursionPolicy`, `RecursionStack`,
`SubAgentDepth`.
- Acceptance: there is one authoritative recursion cap and one error shape,
with compatibility conversion for existing UI/JSON-RPC responses.
- [~] Keep OpenHuman's sub-agent build pipeline as product policy, but move the
pipeline mechanics into TinyAgents graph/sub-agent primitives.
- Product-owned pieces: agent definition resolution, prompt assembly, memory
context, worker-thread mirroring, handoff cache, tool filtering, provider
routing, sandbox scope.
- TinyAgents components to adopt beneath it: `SubAgentSession`,
`SubAgentTool`, `SubAgentNode`, `SubAgentPolicy`, `ToolExecutionContext`,
event lineage, graph status, cancellation, usage/cost rollup.
- Acceptance: use TinyAgents for execution, transcript/session tracking,
graph structure, and lineage without flattening OpenHuman's agent registry
semantics into generic SDK defaults.
## Phase 8 - Registry And Capability Catalog
- [ ] Build a `CapabilityRegistry` projection from OpenHuman registries.
- OpenHuman files: `agent_registry`, `tools`, `mcp_registry`, `inference`,
`cost`, `approval`, `security`, controller registry in `src/core/all.rs`.
- TinyAgents components: `CapabilityRegistry`, `ComponentId`,
`ComponentKind`, model catalog, graph/tool/agent/store/middleware entries.
- Acceptance: OpenHuman models, tools, agents, graphs, stores, and middleware
can be looked up through one policy-aware capability projection.
- [ ] Add registry diagnostics for duplicate names and unsafe aliases.
- OpenHuman files: `src/openhuman/tools/generated.rs`,
`mcp_registry`, `composio`, `agent_registry`.
- TinyAgents components: component names, `ToolRegistry`, diagnostics.
- Acceptance: duplicate tool names, provider-specific aliases, MCP names, and
generated tool names fail closed before model dispatch.
- [ ] Use TinyAgents model catalog shape for local provider catalog snapshots.
- OpenHuman files: `docs/inference-provider-catalog.md`,
`src/openhuman/inference/presets.rs`,
`src/openhuman/inference/provider/factory.rs`.
- TinyAgents components: registry model catalog, local snapshots, price and
context metadata.
- Acceptance: model picker, router, budget estimator, and capability filter
read one normalized catalog projection.
## Phase 9 - Memory, Retrieval, Embeddings, Context, And Cache
- [ ] Adapt OpenHuman memory/retrieval to TinyAgents retriever interfaces.
- OpenHuman files: `memory`, `memory_search`, `memory_tree`,
`agent_memory/memory_loader.rs`, `context/*`.
- TinyAgents components: `EmbeddingModel`, `Retriever`, `VectorStore`,
`ScoredDoc`, context events.
- Acceptance: the agent harness can load retrieval context through a
TinyAgents retriever facade while OpenHuman stores remain authoritative.
- [ ] Move context compaction provenance into TinyAgents events.
- OpenHuman files: `context/README.md`, `tinyagents/summarize.rs`,
`tinyagents/payload_summarizer.rs`.
- TinyAgents components: `SummaryRecord`, `Compressed` events,
`PromptCacheLayout`, cache layout events.
- Acceptance: every summary records source ids, before/after token estimates,
policy version, and whether stable prompt prefix was preserved.
- [ ] Normalize embedding usage/cost records.
- OpenHuman files: `embeddings`, `memory_sync`, `cost`.
- TinyAgents components: embedding usage fields, model catalog pricing.
- Acceptance: embedding calls contribute usage and cost with provider/model,
dimensions, vector count, and source.
## Phase 10 - Dead Code And TinyAgents Re-Expression Audit
Do not delete these blindly. Treat this as an audit list for code that is dead,
vestigial, or generic runtime behavior now expressible through TinyAgents
harness/graph primitives. Delete only after call-site search, compatibility
assessment, and migration coverage are complete.
- [ ] Audit `src/openhuman/agent/harness/engine/*`.
- Current role: surviving seams from the retired in-house turn loops:
`CheckpointStrategy`, `ProgressReporter`, and `TurnProgress`.
- TinyAgents expression: `RunPolicy`, `AgentEvent`, `GraphEvent`,
`EventSink`, `HarnessStatusStore`, cap/stop middleware.
- Candidate outcome: move max-iteration and progress projection into
TinyAgents middleware/events, then delete `engine/*` if no product-specific
compatibility seam remains.
- [ ] Audit `src/openhuman/agent/harness/agent_graph.rs` and built-in
`agent_registry/agents/*/graph.rs` default selectors.
- Current role: per-agent graph hook, but most built-ins return
`AgentGraph::Default`.
- TinyAgents expression: a registry of `CompiledGraph`/graph factories keyed
by agent id, with default graph supplied by the runtime and custom graphs
registered only where they differ.
- Candidate outcome: replace dozens of boilerplate default `graph.rs` files
with registry defaults, keeping files only for agents with custom graphs.
- [x] Audit stale architecture references to removed in-house graph/loop code.
- Current files: `gitbooks/developing/architecture/agent-harness.md`,
`src/openhuman/context/README.md`.
- Candidate stale names: `src/openhuman/agent_graph/`, `GraphBlueprint`,
`run_turn_engine`, `run_tool_call_loop`, `harness/tool_loop.rs`, old
context summarizer files.
- TinyAgents expression: link to live `src/openhuman/tinyagents/*`,
`GraphBuilder`, `AgentHarness`, `ContextCompressionMiddleware`, and
graph-export/status surfaces.
- Candidate outcome: move historical details into a short "pre-migration
history" appendix or remove them from active architecture docs.
- **Partial:** flagged the `agent-harness.md` `agent_graph`/`GraphBlueprint`
section as HISTORICAL-removed (strong inline callout pointing at the live
tinyagents surfaces); fixed `context/README.md` "Used by" line that still
referenced the deleted `reduce_before_call`/`ProviderSummarizer`/
`SegmentRecapSummarizer`/`unified_compaction_enabled`. **Sweep completed:**
every code doc-comment that described `run_turn_engine`/`run_tool_call_loop`/
`tool_loop.rs` as *current* behavior (≈20 sites across 15 files: tools,
security, tokenjuice, triage, orchestration steering, task-local contexts,
cron, host_runtime, event-bus example, test-file headers) now points at the
live tinyagents surfaces; intentionally-historical "legacy X was removed /
parity with" notes were kept. Domain READMEs (tools, tokenjuice, approval)
fixed too — the tokenjuice one now records that `compact_tool_output` lost
its only production caller with the retired loop (re-wiring it as an
`after_tool` middleware; the stale default-Full wrapper is now deleted.
- [ ] Audit `src/openhuman/context/{pipeline,guard,microcompact}.rs`.
- Current role: context stats/session-memory bookkeeping plus older
compaction concepts; live history reduction moved to
`ContextCompressionMiddleware` and `MessageTrimMiddleware`.
- TinyAgents expression: context-window middleware, `Compressed` events,
`PromptCacheLayout`, cache layout events, `SummaryRecord`,
usage/context pressure status.
- Candidate outcome: keep only stats/session-memory state that remains
OpenHuman-specific; move compression policy/provenance into TinyAgents
middleware and delete unused reduction paths.
- [ ] Audit `src/openhuman/tinyagents/payload_summarizer.rs`.
- Current role: oversized tool-result compression via a `summarizer`
sub-agent with a local circuit breaker.
- TinyAgents expression: `ToolMiddleware::after_tool`,
`ContextCompressionMiddleware`, `SummaryRecord`, tool artifact/result
compaction events.
- Candidate outcome: convert to middleware over TinyAgents tool results so
the summarizer is no longer a separate OpenHuman-only hook.
- [ ] Audit `src/openhuman/agent_orchestration/running_subagents.rs`.
- Current role: bespoke live-task registry layered on TinyAgents
`InMemoryTaskStore`, plus watch channels, abort handles, tombstones,
task/session lookup, wait, steer, and cancel operations.
- TinyAgents expression: `TaskStore`, `SteeringCommand`, `SteeringHandle`,
`CancellationToken`, run tree/status store, durable OpenHuman ledger
projections.
- Candidate outcome: keep OpenHuman durable session/worker-thread records,
but collapse transient lifecycle mechanics into TinyAgents task/status
primitives once they can represent wait/steer/hard-abort needs.
- [ ] Audit `src/openhuman/agent_orchestration/tools/spawn_parallel_agents.rs`
after the graph-tool migration.
- Current role: validation, preflight, fanout, worktree setup, event
projection, overlap detection, result formatting.
- TinyAgents expression: graph nodes (`validate`, `dispatch`, `worker`,
`collect`, `finalize`) with `Send` fanout and reducers.
- Candidate outcome: keep a thin tool wrapper that invokes the graph and
formats the existing JSON payload; move orchestration mechanics into the
graph module.
- [ ] Audit hand-rolled `join_all` fanouts that are workflow orchestration, not
simple IO batching.
- Candidate files from current search: `src/openhuman/mcp_registry/registry.rs`,
`src/openhuman/learning/reflection.rs`,
`src/openhuman/inference/local/service/ollama_admin/diagnostics.rs`.
- TinyAgents expression: only migrate fanouts that need agent/run lineage,
checkpointing, policy, cancellation, or graph observability. Leave simple
independent IO probes as ordinary Rust concurrency.
- Candidate outcome: document why each fanout stays as `join_all` or move it
to `graph::parallel::map_reduce` / graph `Send`.
- [x] Audit tool registry comments and docs that still describe retired
direct-loop behavior.
- Current files: `src/openhuman/tools/traits.rs`,
`src/openhuman/tools/README.md`,
`src/openhuman/agent/harness/session/turn/tools.rs`.
- TinyAgents expression: `ToolRegistry`, `ToolMiddleware`, graph tool nodes,
tool safety metadata, `ToolExecutionContext`.
- Candidate outcome: update comments to describe the TinyAgents execution
path and delete references to the retired serial `harness::tool_loop`
dispatcher once no code path uses it.
- **Done:** `tools/traits.rs` concurrency note and `tools/README.md` "Used
by" now describe the tinyagents execution path (`SharedToolAdapter` /
`ToolPolicyMiddleware`); `session/turn/tools.rs` had no stale references.
## Phase 11 - Testing And Conformance
- [ ] Create a focused parity test matrix for the current TinyAgents route.
- OpenHuman files: `src/openhuman/tinyagents/tests.rs`,
`tests/agent_harness_e2e.rs`, `tests/agent_tool_loop_raw_coverage_e2e.rs`.
- TinyAgents components: `AgentHarness`, `AgentEvent`, `ToolRegistry`,
`ModelRegistry`, `MessageTrimMiddleware`, `ContextCompressionMiddleware`.
- Acceptance: chat turn, channel turn, sub-agent turn, unknown tool recovery,
approval denial, streaming text, streaming tool args, reasoning deltas,
early-exit pause, model-call cap, and cost footer all have tests on the
TinyAgents path.
- [x] Add a TinyAgents adapter inventory test.
- OpenHuman files: `src/openhuman/tinyagents/mod.rs`,
`src/openhuman/tinyagents/tests.rs`.
- Acceptance: one test asserts that the shared runner registers model, tools,
middleware, event bridge, context compression, stop hooks, and unknown-tool
policy in the intended order.
- **Done:** harness assembly extracted from `run_turn_via_tinyagents_shared`
into a testable `assemble_turn_harness` returning `AssembledTurnHarness`
(harness + cursor/error-slot/halt-summary/outcome-sink/steering/early-exit
seams). `adapter_inventory_registers_model_tools_and_middleware` asserts
model registry, callable tools + unknown-tool policy, 9 lifecycle + 2
around-tool middlewares, steering handle, early-exit hook;
`adapter_inventory_gates_context_middleware_on_window` proves the
compression/trim gating. SDK gap: `MiddlewareStack` exposes lengths but not
names, so exact ordering is documented at the registration sites and
guarded by counts.
- [ ] Port behavior clusters to TinyAgents testkit.
- OpenHuman files: `src/openhuman/agent/harness/*_tests.rs`,
`tests/agent_*`.
- TinyAgents components: harness `testkit`, graph `assert_graph`,
`GraphEventRecorder`, mock models/tools.
- Acceptance: legacy assertions over loop wording are replaced by assertions
over TinyAgents events, checkpoints, graph metadata, and final transcript.
- [ ] Add cross-module e2e tests for graph/sub-agent/model/tool composition.
- Candidate tests: workflow run with child sub-agents, delegation with review
loop, council fanout, MCP tool call, Composio approval denial, memory
retrieval plus summarization.
- Acceptance: tests cover default features, `openai`-feature compatible code
paths where available, and all-features where dependency constraints allow.
- [ ] Add fuzz-style graph composition tests.
- TinyAgents components: graph testkit, reducers, command routing, subgraph
nodes, sub-agent fake nodes.
- Acceptance: generated small graphs cover direct edges, conditional routes,
`Send` fanout, joins, interrupts, recursion caps, and checkpoint resume.
## Suggested Execution Order
1. Confirm version/feature compatibility and SDK-extension gaps.
2. Move tool safety/approval/security into TinyAgents middleware.
3. Normalize usage/cost records and parent/child rollups.
4. Persist TinyAgents event/status journals to the OpenHuman run ledger.
5. Build the OpenHuman `CapabilityRegistry` projection.
6. Convert one high-value built-in agent to a bespoke TinyAgents graph.
7. Adapt memory/retrieval/context surfaces where the TinyAgents interfaces are
a good fit.
8. Audit and remove or re-express dead/vestigial runtime code through
TinyAgents harness/graph primitives.
9. Finish with parity, adapter-inventory, conformance, e2e, and fuzz-style tests.
## Non-Goals For The First Migration Wave
- Do not replace OpenHuman's durable run ledgers with TinyAgents SQLite storage
until the one-time
transcript/session migration, TinyAgents store/status adapter, and
restart/resume parity tests are complete.
- Do not expose TinyAgents' OpenAI provider directly to product code while
OpenHuman provider config, credentials, OAuth, and billing classification are
still the product source of truth.
- Do not remove `DomainEvent` until all existing subscribers have a TinyAgents
event/status replacement.