Capabilities

Capabilities

An index of what each TinyAgents crate exposes, with a link to the page that documents it. Start here, find the crate you need, and follow the link.

The workspace has no facade crate: Harness, Graph runtime, Registry, the task and orchestration crates (tinyagents-tasks, tinyagents-orchestration), session, runtime, definition and live are separate crates, plus Providers at the leaves and testkit/conformance helpers for tests. See Architecture for how they depend on each other.

Harness — model calls, tools, control

Capability	What it does	Where
Agent loop	Default model→tool→model loop with run identity + limits	Harness
Provider-neutral model calls	One ChatModel trait; ModelRequest/ModelResponse across hosts	Harness
Model resolution & selection	ModelRegistry resolves a call via ModelSelection/capabilities	Harness
Typed tools	Tool trait with JSON schemas and a ToolRegistry; nested call_tool (off by default)	Harness
Middleware hooks	before/after hooks around agent, model, and tool calls	Harness
Structured output	Extract a typed value via a provider-appropriate strategy	Harness
Streaming	Model-delta and event streaming through the loop	Harness
Usage & cost	Token accounting and pricing roll-ups across the run tree	Harness
Limits, retry, failover	Fail-closed caps, backoff, reason-aware failover, rate limiting	Harness
Response & prompt cache	Local response cache + provider prompt/KV-cache layout	Harness
Memory & retrieval	Host-supplied AgentMemory, the Retriever contract, embeddings via tinyinference-embeddings	Harness
Sub-agents (agents as tools)	tinyagents-orchestration: SubagentDriver, spawn admission, job registry, completion routing	Harness
Steering	Typed runtime control (pause/resume/redirect, opt-in SwitchModel) of a running agent	Harness
Summarization	Context-window-aware transcript compaction	Harness
Events & run status	In-memory event spine (TurnStarted/TurnCompleted/MessageAppended, tool progress) + run status + typed TerminalOutcome	Harness
Durable observability	Journals, status stores, sinks, latency metrics	Harness
Testkit	Fakes, recorders, trajectory assertions	Harness
Loop detection	Staged nudge/warn/block/halt on stalled or repeating tool calls	Harness

Graph runtime — durable typed workflows

Capability	What it does	Where
Nodes & edges	Named nodes wired by static/conditional edges over typed state	Graph Runtime
Dynamic routing	Command.goto + routing precedence resolve next targets	Graph Runtime
Send fan-out (map-reduce)	Schedule nodes with per-invocation args for the map step	Graph Runtime
Reducers & channels	Merge branch updates deterministically; channel-per-field state	Graph Runtime
Parallel fan-out	Run multi-node supersteps concurrently, folded in index order	Graph Runtime
Checkpoints & durability	Persist at superstep boundaries; in-memory/file/SQLite backends	Graph Runtime
Time travel	Read/fork/update state history against a checkpointer	Graph Runtime
Interrupts & resume	Human-in-the-loop pause and resume with a payload	Graph Runtime
Subgraphs	Embed a compiled graph as a node	Graph Runtime
Sub-agent nodes (deprecated)	Embed a harness agent as a graph node; use SubagentDriver instead	Graph Runtime
Recursion policy	Bound depth, per-node visits, and total steps for sub-agent/subgraph chains	Graph Runtime
Orchestration tools	Model-callable child-task supervision (orchestrate_*), task stores, detached-task registry, completion router; now in tinyagents-tasks	Graph Runtime
Goals and todos	A durable ThreadGoal and a TodoList checklist, exposed as harness tools	Goals and Todos
Streaming & events	GraphEvents + StreamMode projections	Graph Runtime
Durable graph observability	Journals, status store, latency metrics	Graph Runtime
Topology export	JSON/Mermaid export of a graph's structure	Graph Runtime
Graph testkit	Deterministic node doubles + graph assertions	Graph Runtime

Registry — named capability catalog

Capability	What it does	Where
Component identity	ComponentId/ComponentKind/ComponentMetadata describe capabilities	Registry
Capability registry	Register/resolve models, tools, agent definitions, capability bundles by name	Registry
Model catalog	Offline pricing, context windows, capability facts	Registry
Model router	Declarative workload tiers mapped onto registered models with fallback chains	Registry

Providers — model backends at the leaves

Capability	What it does	Where
Offline mock model	Deterministic, network-free default build	Providers
OpenAI + compatible hosts	One adapter for OpenAI and OpenAI-compatible endpoints (Anthropic compat, Ollama, DeepSeek, Groq, xAI, OpenRouter, Together, Mistral)	Providers
List available models	OpenAiModel::list_models() — runtime model discovery via GET /models	Providers
Provider inference	Resolve a provider from a model string	Providers
Capability profiles	Reject impossible requests; pick capability-satisfying fallbacks	Providers

Session — durable history

Capability	What it does	Where
Session and message history	SQLite-backed (WAL + FTS5) sessions, messages, tool calls, cost metadata, parent/child lineage	tinyagents-session crate docs (crates/tinyagents-session/README.md)
Run ledger	Durable state for background agent/workflow execution	tinyagents-session crate docs

Testing — testkit & conformance

Capability	What it does	Where
Harness testkit	Scripted models, fake tools, trajectory asserts	Harness
Graph testkit	Node doubles + fluent graph assertions	Graph Runtime
Storage conformance	Reusable contracts for task stores and checkpointers	Graph Runtime

See also

• Home — orientation and the crate overview.
• Examples — runnable end-to-end demonstrations.
• Architecture — how the crates compose.