selectools 0.25.1__tar.gz → 0.27.0__tar.gz

{selectools-0.25.1/src/selectools.egg-info → selectools-0.27.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: selectools
-Version: 0.25.1
+Version: 0.27.0
 Summary: Production-ready Python framework for AI agents with multi-agent graphs, hybrid RAG, guardrails, audit logging, 50 evaluators, and a visual builder. Supports OpenAI, Anthropic, Gemini, Ollama. By NichevLabs.
 Author-email: John Nichev <johnnichev@gmail.com>
 Maintainer-email: NichevLabs <support@nichevlabs.com>
@@ -59,6 +59,12 @@ Provides-Extra: toolbox
 Requires-Dist: requests>=2.28.0; extra == "toolbox"
 Requires-Dist: slack-sdk>=3.27.0; extra == "toolbox"
 Requires-Dist: pdfplumber>=0.11.0; extra == "toolbox"
+Provides-Extra: aws
+Requires-Dist: boto3>=1.34.0; extra == "aws"
+Provides-Extra: browser
+Requires-Dist: playwright>=1.40.0; extra == "browser"
+Provides-Extra: mongo
+Requires-Dist: pymongo>=4.0.0; extra == "mongo"
 Provides-Extra: evals
 Requires-Dist: pyyaml>=6.0.0; extra == "evals"
 Provides-Extra: mcp
@@ -108,6 +114,37 @@ result = AgentGraph.chain(planner, writer, reviewer).run("Write a blog post")
 # selectools serve agent.yaml
 ```
 
+## What's New in v0.27
+
+### v0.27.0 — Scheduling, Reasoning & New Backends
+
+A feature release landing the post-1.0 backlog: scheduled agents, composable reasoning, two new session backends, an injection guardrail, and the rest of the v1.1 candidate set. All additive (`@beta`); no breaking changes.
+
+- **Scheduled agents** — `AgentScheduler` runs an agent on a `cron("0 9 * * *")` or `every(minutes=5)` schedule. Async loop that sleeps until the next due job, per-job `max_runs` / `on_result` / `start_immediately`, failure isolation, and an injectable clock for tests. Stdlib-only (`selectools.scheduler`).
+- **Reasoning tools** — `make_reasoning_tools(min_steps, max_steps)` adds `think` / `analyze` tools that make the agent's reasoning explicit, inspectable steps (vs the passive `reasoning_strategy` prompt). `max_steps` is enforced (a real guard against reasoning loops); `min_steps` is guidance.
+- **Two new session backends** — `MongoSessionStore` and `DynamoDBSessionStore` bring the `SessionStore` count to six. Full protocol (save/load/list/delete/exists/branch/search), namespace isolation, and optional server-side TTL. `pip install selectools[mongo]` / `selectools[aws]`.
+- **Prompt-injection guardrail** — `PromptInjectionGuardrail` catches templated jailbreak/injection attacks ("ignore previous instructions", "reveal your system prompt", `<system>`/`[INST]` spoofing, "developer mode"/"DAN") with high-precision patterns. Heuristic tier — no model hosting required.
+- **Agentic memory completed** — the auto-injected `recall` tool joins `remember`, so the agent can both store and query its `KnowledgeMemory` on demand.
+- **UnifiedMemory via config** — `AgentConfig(memory=MemoryConfig(unified=True, ...))` makes the tiered `UnifiedMemory` (STM→LTM auto-promotion, entity + episodic tiers, token-aware compaction) reachable from config; default off.
+- **Toolbox 48 → 56 tools** — added Discord, AWS S3, headless browser (Playwright), and image generation (OpenAI Images) categories.
+- **Cache-rate cost for OpenAI + Gemini** — `calculate_cost_with_cached_input()` prices cached prompt tokens at each provider's published rate (`cached_prompt_cost` on `ModelInfo`); 24 rates source-verified.
+- **Performance benchmarks published** — framework overhead measured and documented (`docs/modules/BENCHMARKS.md`).
+
+See `CHANGELOG.md` for the full entry (7,700+ tests, 115 examples, 115 models).
+
+## What's New in v0.26
+
+### v0.26.0 — Safety Patch & Verified Registry
+
+A mid-bake safety patch plus a fully source-verified model registry.
+
+- **Confirm parser safety fix** — `RegexConfirmParser` treated replies like "se você não pode apagar, tudo bem" / "tú no puedes borrar" as confirmations: the restated-verb branch matched the action verb without checking for negation elsewhere in the reply. A negation token anywhere now vetoes that branch; bare confirms are unaffected. **All `selectools.pending` consumers should upgrade immediately.**
+- **Model registry refresh** — 152 → 115 models, every entry source-verified: claude-fable-5 / opus-4-8 / opus-4-7, the gpt-5.5 family, and the gemini-3.5 line added; retired and shut-down models removed (**BREAKING** if you referenced their registry constants — e.g. `AnthropicModels.CLAUDE_3_5_SONNET` — but those IDs 404 at their providers regardless); `claude-opus-4-1` mispricing corrected ($5/$25 → $15/$75); gpt-5 family context specs fixed (400k/128k); `GeminiEmbeddingProvider` default moved off the retired `text-embedding-004` to `gemini-embedding-001`.
+- **Cache-aware cost calculation** — `calculate_cost(..., cache_read_input_tokens=..., cache_creation_input_tokens=...)` prices prompt-cache traffic at Anthropic's published rates (reads 0.1×, 5-min writes 1.25×); `AnthropicProvider`'s `UsageStats.cost_usd` is now cache-accurate.
+- **Fixes** — A2A returns JSON-RPC -32602 (not HTTP 500) for non-dict `message.parts` elements; Gemini embedding dimension constant corrected to the actual 3072 for `gemini-embedding-001`/`-2` (vectors themselves unchanged); bake-hunt test module no longer requires optional extras in CI.
+
+See `CHANGELOG.md` for the full entry (7,420 tests, 111 examples, 115 models).
+
 ## What's New in v0.25
 
 ### v0.25.0 — Hardening & v1.0 Prep
@@ -618,11 +655,13 @@ report.to_html("report.html")
 | **Dynamic Tools** | Load tools from files/directories at runtime. Add, remove, replace tools without restarting. |
 | **Response Caching** | LRU + TTL in-memory cache and Redis backend. Avoid redundant LLM calls for identical requests. |
 | **Routing Mode** | Agent selects a tool without executing it. Use for intent classification and request routing. |
-| **Guardrails Engine** | Input/output validation pipeline with PII redaction, topic blocking, toxicity detection, and format enforcement. |
+| **Guardrails Engine** | Input/output validation pipeline with PII redaction, prompt-injection detection, topic blocking, toxicity detection, and format enforcement. |
 | **Audit Logging** | JSONL audit trail with privacy controls (redact, hash, omit) and daily rotation. |
 | **Tool Output Screening** | Prompt injection detection with 15 built-in patterns. Per-tool or global. |
 | **Coherence Checking** | LLM-based verification that tool calls match user intent — catches injection-driven tool misuse. |
-| **Persistent Sessions** | `SessionStore` with JSON file, SQLite, Redis, and Supabase backends. Auto-save/load with TTL expiry, cross-session `search()`. |
+| **Persistent Sessions** | `SessionStore` with JSON file, SQLite, Redis, Supabase, MongoDB, and DynamoDB backends. Auto-save/load with TTL expiry, cross-session `search()`. |
+| **Scheduled Agents** | `AgentScheduler` runs an agent on a `cron()` or `every()` schedule — async loop, per-job `max_runs`, failure isolation, injectable clock. |
+| **Reasoning Tools** | `make_reasoning_tools()` adds `think`/`analyze` tools that make reasoning explicit and inspectable, bounded by `min_steps`/`max_steps`. |
 | **Agent-as-API** | `AgentAPI` serves any agent as a Starlette REST API — chat, SSE streaming, session CRUD, bearer auth, per-user session isolation. |
 | **A2A Protocol** | `A2AServer` + `A2AClient`: Agent Card discovery and JSON-RPC 2.0 messaging between agents. |
 | **LiteLLM Bridge** | `LiteLLMProvider` unlocks 100+ models (DeepSeek, Groq, Bedrock, ...) through one provider class. |
@@ -633,7 +672,7 @@ report.to_html("report.html")
 | **Prompt Caching** | Anthropic `cache_system`/`cache_tools` markers with cache hit-rate fields on `UsageStats`. |
 | **Entity Memory** | LLM-based entity extraction with deduplication, LRU pruning, and system prompt injection. |
 | **Knowledge Graph** | Relationship triple extraction with in-memory and SQLite storage and keyword-based querying. |
-| **Cross-Session Knowledge** | Daily logs + persistent facts with auto-registered `remember` tool. |
+| **Cross-Session Knowledge** | Daily logs + persistent facts with auto-registered `remember` + `recall` tools. |
 | **MCP Integration** | Connect to any MCP tool server (stdio + HTTP). MCPClient, MultiMCPClient, MCPServer. Circuit breaker, retry, graceful degradation. |
 | **Eval Framework** | 50 built-in evaluators (30 deterministic + 21 LLM-as-judge). A/B testing, regression detection, snapshot testing, HTML reports, JUnit XML, CI integration. |
 | **Multi-Agent Orchestration** | `AgentGraph` for directed agent graphs, `SupervisorAgent` with 4 strategies, HITL via generator nodes, parallel execution, checkpointing, subgraph composition. |
@@ -659,9 +698,12 @@ report.to_html("report.html")
 - **Advanced Chunking**: Semantic + contextual chunking for better retrieval
 - **Dynamic Tool Loading**: Plugin system with hot-reload support
 - **Response Caching**: InMemoryCache and RedisCache with stats tracking
-- **152 Model Registry**: Type-safe constants with pricing and metadata
-- **Pre-built Toolbox**: 48 tools for files, data, text, datetime, web, code, search, GitHub, DB, calculator, email, PDF, Slack, Notion, Linear
-- **Persistent Sessions**: 4 backends (JSON file, SQLite, Redis, Supabase) with TTL and cross-session search
+- **115 Model Registry**: Type-safe constants with pricing and metadata
+- **Pre-built Toolbox**: 56 tools for files, data, text, datetime, web, code, search, GitHub, DB, calculator, email, PDF, Slack, Notion, Linear, Discord, S3, browser, image gen
+- **Persistent Sessions**: 6 backends (JSON file, SQLite, Redis, Supabase, MongoDB, DynamoDB) with TTL and cross-session search
+- **Scheduled Agents**: `AgentScheduler` runs an agent on a cron or interval schedule with per-job max-runs, failure isolation, and an async loop
+- **Reasoning Tools**: `think`/`analyze` tools make reasoning explicit, inspectable, and bounded by min/max steps
+- **Prompt-Injection Guardrail**: `PromptInjectionGuardrail` — heuristic jailbreak/injection detection with high-precision patterns
 - **Entity Memory**: LLM-based named entity extraction and tracking
 - **Unified Memory**: tiered conversation/knowledge/entity/episodic memory with token-aware compaction
 - **Knowledge Backends**: Supabase/Redis blob persistence for KnowledgeMemory on ephemeral infra
@@ -669,7 +711,7 @@ report.to_html("report.html")
 - **Deferred Confirmation**: `selectools.pending` for chat-channel destructive-tool confirmation
 - **Anthropic Prompt Caching**: `cache_system`/`cache_tools` with hit-rate visibility on `UsageStats`
 - **Knowledge Graph**: Triple extraction with in-memory and SQLite storage
-- **Cross-Session Knowledge**: Daily logs + persistent memory with `remember` tool, pluggable stores (File, SQLite), importance scoring, TTL
+- **Cross-Session Knowledge**: Daily logs + persistent memory with auto-injected `remember` + `recall` tools, pluggable stores (File, SQLite), importance scoring, TTL
 - **Token Budget & Cancellation**: `max_total_tokens`, `max_cost_usd` hard limits; `CancellationToken` for cooperative stopping
 - **Token Estimation**: `estimate_run_tokens()` for pre-execution budget checks
 - **Model Switching**: `model_selector` callback for per-iteration model selection
@@ -678,10 +720,10 @@ report.to_html("report.html")
 - **Conversation Branching**: `ConversationMemory.branch()` and `SessionStore.branch()` for A/B exploration and checkpointing
 - **Multi-Agent Orchestration**: `AgentGraph` with routing, parallel execution, HITL, checkpointing; `SupervisorAgent` with 4 strategies (plan_and_execute, round_robin, dynamic, magentic)
 - **Composable Pipelines**: `Pipeline` + `@step` + `|` operator + `parallel()` + `branch()` — chain agents, tools, and transforms
-- **111 Examples**: Multi-agent graphs, RAG, hybrid search, streaming, structured output, traces, batch, policy, observer, guardrails, audit, sessions (incl. Supabase), entity memory, knowledge graph, eval framework, advanced agent patterns, stability markers, HTML trace viewer, agent-as-API, A2A, routing, unified memory, and more
+- **115 Examples**: Multi-agent graphs, RAG, hybrid search, streaming, structured output, traces, batch, policy, observer, guardrails, audit, sessions (incl. Supabase), entity memory, knowledge graph, eval framework, advanced agent patterns, stability markers, HTML trace viewer, agent-as-API, A2A, routing, unified memory, scheduled agents, reasoning tools, and more
 - **Built-in Eval Framework**: 50 evaluators (30 deterministic + 21 LLM-as-judge), A/B testing, regression detection, HTML reports, JUnit XML, snapshot testing
 - **AgentObserver Protocol**: 46 lifecycle events with `run_id` correlation, `LoggingObserver`, `SimpleStepObserver`, OTel export
-- **7268 Tests**: Unit, integration, regression, and E2E with real API calls
+- **7700+ Tests**: Unit, integration, regression, and E2E with real API calls
 
 ## Install
 
@@ -1035,7 +1077,7 @@ See [docs/modules/STREAMING.md](https://github.com/johnnichev/selectools/blob/ma
 ```python
 from selectools.models import OpenAI, Anthropic, Gemini, Ollama
 
-# IDE autocomplete for all 152 models with pricing metadata
+# IDE autocomplete for all 115 models with pricing metadata
 model = OpenAI.GPT_4O_MINI
 print(f"Cost: ${model.prompt_cost}/${model.completion_cost} per 1M tokens")
 print(f"Context: {model.context_window:,} tokens")
@@ -1183,7 +1225,7 @@ Examples are numbered by difficulty. Start from 01 and work your way up.
 |---|---|---|---|
 | 01 | `01_hello_world.py` | First agent, `@tool`, `ask()` | No |
 | 02 | `02_search_weather.py` | ToolRegistry, multiple tools | No |
-| 03 | `03_toolbox.py` | 48 pre-built tools (file, data, text, datetime, web, code, search, and more) | No |
+| 03 | `03_toolbox.py` | 56 pre-built tools (file, data, text, datetime, web, code, search, and more) | No |
 | 04 | `04_conversation_memory.py` | Multi-turn memory | Yes |
 | 05 | `05_cost_tracking.py` | Token counting, cost warnings | Yes |
 | 06 | `06_async_agent.py` | `arun()`, concurrent agents, FastAPI | Yes |

{selectools-0.25.1 → selectools-0.27.0}/README.md

@@ -30,6 +30,37 @@ result = AgentGraph.chain(planner, writer, reviewer).run("Write a blog post")
 # selectools serve agent.yaml
 ```
 
+## What's New in v0.27
+
+### v0.27.0 — Scheduling, Reasoning & New Backends
+
+A feature release landing the post-1.0 backlog: scheduled agents, composable reasoning, two new session backends, an injection guardrail, and the rest of the v1.1 candidate set. All additive (`@beta`); no breaking changes.
+
+- **Scheduled agents** — `AgentScheduler` runs an agent on a `cron("0 9 * * *")` or `every(minutes=5)` schedule. Async loop that sleeps until the next due job, per-job `max_runs` / `on_result` / `start_immediately`, failure isolation, and an injectable clock for tests. Stdlib-only (`selectools.scheduler`).
+- **Reasoning tools** — `make_reasoning_tools(min_steps, max_steps)` adds `think` / `analyze` tools that make the agent's reasoning explicit, inspectable steps (vs the passive `reasoning_strategy` prompt). `max_steps` is enforced (a real guard against reasoning loops); `min_steps` is guidance.
+- **Two new session backends** — `MongoSessionStore` and `DynamoDBSessionStore` bring the `SessionStore` count to six. Full protocol (save/load/list/delete/exists/branch/search), namespace isolation, and optional server-side TTL. `pip install selectools[mongo]` / `selectools[aws]`.
+- **Prompt-injection guardrail** — `PromptInjectionGuardrail` catches templated jailbreak/injection attacks ("ignore previous instructions", "reveal your system prompt", `<system>`/`[INST]` spoofing, "developer mode"/"DAN") with high-precision patterns. Heuristic tier — no model hosting required.
+- **Agentic memory completed** — the auto-injected `recall` tool joins `remember`, so the agent can both store and query its `KnowledgeMemory` on demand.
+- **UnifiedMemory via config** — `AgentConfig(memory=MemoryConfig(unified=True, ...))` makes the tiered `UnifiedMemory` (STM→LTM auto-promotion, entity + episodic tiers, token-aware compaction) reachable from config; default off.
+- **Toolbox 48 → 56 tools** — added Discord, AWS S3, headless browser (Playwright), and image generation (OpenAI Images) categories.
+- **Cache-rate cost for OpenAI + Gemini** — `calculate_cost_with_cached_input()` prices cached prompt tokens at each provider's published rate (`cached_prompt_cost` on `ModelInfo`); 24 rates source-verified.
+- **Performance benchmarks published** — framework overhead measured and documented (`docs/modules/BENCHMARKS.md`).
+
+See `CHANGELOG.md` for the full entry (7,700+ tests, 115 examples, 115 models).
+
+## What's New in v0.26
+
+### v0.26.0 — Safety Patch & Verified Registry
+
+A mid-bake safety patch plus a fully source-verified model registry.
+
+- **Confirm parser safety fix** — `RegexConfirmParser` treated replies like "se você não pode apagar, tudo bem" / "tú no puedes borrar" as confirmations: the restated-verb branch matched the action verb without checking for negation elsewhere in the reply. A negation token anywhere now vetoes that branch; bare confirms are unaffected. **All `selectools.pending` consumers should upgrade immediately.**
+- **Model registry refresh** — 152 → 115 models, every entry source-verified: claude-fable-5 / opus-4-8 / opus-4-7, the gpt-5.5 family, and the gemini-3.5 line added; retired and shut-down models removed (**BREAKING** if you referenced their registry constants — e.g. `AnthropicModels.CLAUDE_3_5_SONNET` — but those IDs 404 at their providers regardless); `claude-opus-4-1` mispricing corrected ($5/$25 → $15/$75); gpt-5 family context specs fixed (400k/128k); `GeminiEmbeddingProvider` default moved off the retired `text-embedding-004` to `gemini-embedding-001`.
+- **Cache-aware cost calculation** — `calculate_cost(..., cache_read_input_tokens=..., cache_creation_input_tokens=...)` prices prompt-cache traffic at Anthropic's published rates (reads 0.1×, 5-min writes 1.25×); `AnthropicProvider`'s `UsageStats.cost_usd` is now cache-accurate.
+- **Fixes** — A2A returns JSON-RPC -32602 (not HTTP 500) for non-dict `message.parts` elements; Gemini embedding dimension constant corrected to the actual 3072 for `gemini-embedding-001`/`-2` (vectors themselves unchanged); bake-hunt test module no longer requires optional extras in CI.
+
+See `CHANGELOG.md` for the full entry (7,420 tests, 111 examples, 115 models).
+
 ## What's New in v0.25
 
 ### v0.25.0 — Hardening & v1.0 Prep
@@ -540,11 +571,13 @@ report.to_html("report.html")
 | **Dynamic Tools** | Load tools from files/directories at runtime. Add, remove, replace tools without restarting. |
 | **Response Caching** | LRU + TTL in-memory cache and Redis backend. Avoid redundant LLM calls for identical requests. |
 | **Routing Mode** | Agent selects a tool without executing it. Use for intent classification and request routing. |
-| **Guardrails Engine** | Input/output validation pipeline with PII redaction, topic blocking, toxicity detection, and format enforcement. |
+| **Guardrails Engine** | Input/output validation pipeline with PII redaction, prompt-injection detection, topic blocking, toxicity detection, and format enforcement. |
 | **Audit Logging** | JSONL audit trail with privacy controls (redact, hash, omit) and daily rotation. |
 | **Tool Output Screening** | Prompt injection detection with 15 built-in patterns. Per-tool or global. |
 | **Coherence Checking** | LLM-based verification that tool calls match user intent — catches injection-driven tool misuse. |
-| **Persistent Sessions** | `SessionStore` with JSON file, SQLite, Redis, and Supabase backends. Auto-save/load with TTL expiry, cross-session `search()`. |
+| **Persistent Sessions** | `SessionStore` with JSON file, SQLite, Redis, Supabase, MongoDB, and DynamoDB backends. Auto-save/load with TTL expiry, cross-session `search()`. |
+| **Scheduled Agents** | `AgentScheduler` runs an agent on a `cron()` or `every()` schedule — async loop, per-job `max_runs`, failure isolation, injectable clock. |
+| **Reasoning Tools** | `make_reasoning_tools()` adds `think`/`analyze` tools that make reasoning explicit and inspectable, bounded by `min_steps`/`max_steps`. |
 | **Agent-as-API** | `AgentAPI` serves any agent as a Starlette REST API — chat, SSE streaming, session CRUD, bearer auth, per-user session isolation. |
 | **A2A Protocol** | `A2AServer` + `A2AClient`: Agent Card discovery and JSON-RPC 2.0 messaging between agents. |
 | **LiteLLM Bridge** | `LiteLLMProvider` unlocks 100+ models (DeepSeek, Groq, Bedrock, ...) through one provider class. |
@@ -555,7 +588,7 @@ report.to_html("report.html")
 | **Prompt Caching** | Anthropic `cache_system`/`cache_tools` markers with cache hit-rate fields on `UsageStats`. |
 | **Entity Memory** | LLM-based entity extraction with deduplication, LRU pruning, and system prompt injection. |
 | **Knowledge Graph** | Relationship triple extraction with in-memory and SQLite storage and keyword-based querying. |
-| **Cross-Session Knowledge** | Daily logs + persistent facts with auto-registered `remember` tool. |
+| **Cross-Session Knowledge** | Daily logs + persistent facts with auto-registered `remember` + `recall` tools. |
 | **MCP Integration** | Connect to any MCP tool server (stdio + HTTP). MCPClient, MultiMCPClient, MCPServer. Circuit breaker, retry, graceful degradation. |
 | **Eval Framework** | 50 built-in evaluators (30 deterministic + 21 LLM-as-judge). A/B testing, regression detection, snapshot testing, HTML reports, JUnit XML, CI integration. |
 | **Multi-Agent Orchestration** | `AgentGraph` for directed agent graphs, `SupervisorAgent` with 4 strategies, HITL via generator nodes, parallel execution, checkpointing, subgraph composition. |
@@ -581,9 +614,12 @@ report.to_html("report.html")
 - **Advanced Chunking**: Semantic + contextual chunking for better retrieval
 - **Dynamic Tool Loading**: Plugin system with hot-reload support
 - **Response Caching**: InMemoryCache and RedisCache with stats tracking
-- **152 Model Registry**: Type-safe constants with pricing and metadata
-- **Pre-built Toolbox**: 48 tools for files, data, text, datetime, web, code, search, GitHub, DB, calculator, email, PDF, Slack, Notion, Linear
-- **Persistent Sessions**: 4 backends (JSON file, SQLite, Redis, Supabase) with TTL and cross-session search
+- **115 Model Registry**: Type-safe constants with pricing and metadata
+- **Pre-built Toolbox**: 56 tools for files, data, text, datetime, web, code, search, GitHub, DB, calculator, email, PDF, Slack, Notion, Linear, Discord, S3, browser, image gen
+- **Persistent Sessions**: 6 backends (JSON file, SQLite, Redis, Supabase, MongoDB, DynamoDB) with TTL and cross-session search
+- **Scheduled Agents**: `AgentScheduler` runs an agent on a cron or interval schedule with per-job max-runs, failure isolation, and an async loop
+- **Reasoning Tools**: `think`/`analyze` tools make reasoning explicit, inspectable, and bounded by min/max steps
+- **Prompt-Injection Guardrail**: `PromptInjectionGuardrail` — heuristic jailbreak/injection detection with high-precision patterns
 - **Entity Memory**: LLM-based named entity extraction and tracking
 - **Unified Memory**: tiered conversation/knowledge/entity/episodic memory with token-aware compaction
 - **Knowledge Backends**: Supabase/Redis blob persistence for KnowledgeMemory on ephemeral infra
@@ -591,7 +627,7 @@ report.to_html("report.html")
 - **Deferred Confirmation**: `selectools.pending` for chat-channel destructive-tool confirmation
 - **Anthropic Prompt Caching**: `cache_system`/`cache_tools` with hit-rate visibility on `UsageStats`
 - **Knowledge Graph**: Triple extraction with in-memory and SQLite storage
-- **Cross-Session Knowledge**: Daily logs + persistent memory with `remember` tool, pluggable stores (File, SQLite), importance scoring, TTL
+- **Cross-Session Knowledge**: Daily logs + persistent memory with auto-injected `remember` + `recall` tools, pluggable stores (File, SQLite), importance scoring, TTL
 - **Token Budget & Cancellation**: `max_total_tokens`, `max_cost_usd` hard limits; `CancellationToken` for cooperative stopping
 - **Token Estimation**: `estimate_run_tokens()` for pre-execution budget checks
 - **Model Switching**: `model_selector` callback for per-iteration model selection
@@ -600,10 +636,10 @@ report.to_html("report.html")
 - **Conversation Branching**: `ConversationMemory.branch()` and `SessionStore.branch()` for A/B exploration and checkpointing
 - **Multi-Agent Orchestration**: `AgentGraph` with routing, parallel execution, HITL, checkpointing; `SupervisorAgent` with 4 strategies (plan_and_execute, round_robin, dynamic, magentic)
 - **Composable Pipelines**: `Pipeline` + `@step` + `|` operator + `parallel()` + `branch()` — chain agents, tools, and transforms
-- **111 Examples**: Multi-agent graphs, RAG, hybrid search, streaming, structured output, traces, batch, policy, observer, guardrails, audit, sessions (incl. Supabase), entity memory, knowledge graph, eval framework, advanced agent patterns, stability markers, HTML trace viewer, agent-as-API, A2A, routing, unified memory, and more
+- **115 Examples**: Multi-agent graphs, RAG, hybrid search, streaming, structured output, traces, batch, policy, observer, guardrails, audit, sessions (incl. Supabase), entity memory, knowledge graph, eval framework, advanced agent patterns, stability markers, HTML trace viewer, agent-as-API, A2A, routing, unified memory, scheduled agents, reasoning tools, and more
 - **Built-in Eval Framework**: 50 evaluators (30 deterministic + 21 LLM-as-judge), A/B testing, regression detection, HTML reports, JUnit XML, snapshot testing
 - **AgentObserver Protocol**: 46 lifecycle events with `run_id` correlation, `LoggingObserver`, `SimpleStepObserver`, OTel export
-- **7268 Tests**: Unit, integration, regression, and E2E with real API calls
+- **7700+ Tests**: Unit, integration, regression, and E2E with real API calls
 
 ## Install
 
@@ -957,7 +993,7 @@ See [docs/modules/STREAMING.md](https://github.com/johnnichev/selectools/blob/ma
 ```python
 from selectools.models import OpenAI, Anthropic, Gemini, Ollama
 
-# IDE autocomplete for all 152 models with pricing metadata
+# IDE autocomplete for all 115 models with pricing metadata
 model = OpenAI.GPT_4O_MINI
 print(f"Cost: ${model.prompt_cost}/${model.completion_cost} per 1M tokens")
 print(f"Context: {model.context_window:,} tokens")
@@ -1105,7 +1141,7 @@ Examples are numbered by difficulty. Start from 01 and work your way up.
 |---|---|---|---|
 | 01 | `01_hello_world.py` | First agent, `@tool`, `ask()` | No |
 | 02 | `02_search_weather.py` | ToolRegistry, multiple tools | No |
-| 03 | `03_toolbox.py` | 48 pre-built tools (file, data, text, datetime, web, code, search, and more) | No |
+| 03 | `03_toolbox.py` | 56 pre-built tools (file, data, text, datetime, web, code, search, and more) | No |
 | 04 | `04_conversation_memory.py` | Multi-turn memory | Yes |
 | 05 | `05_cost_tracking.py` | Token counting, cost warnings | Yes |
 | 06 | `06_async_agent.py` | `arun()`, concurrent agents, FastAPI | Yes |

{selectools-0.25.1 → selectools-0.27.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "selectools"
-version = "0.25.1"
+version = "0.27.0"
 description = "Production-ready Python framework for AI agents with multi-agent graphs, hybrid RAG, guardrails, audit logging, 50 evaluators, and a visual builder. Supports OpenAI, Anthropic, Gemini, Ollama. By NichevLabs."
 readme = "README.md"
 requires-python = ">=3.9"
@@ -80,6 +80,15 @@ toolbox = [
     "slack-sdk>=3.27.0",
     "pdfplumber>=0.11.0",
 ]
+aws = [
+    "boto3>=1.34.0",
+]
+browser = [
+    "playwright>=1.40.0",
+]
+mongo = [
+    "pymongo>=4.0.0",
+]
 evals = [
     "pyyaml>=6.0.0",
 ]

{selectools-0.25.1 → selectools-0.27.0}/src/selectools/__init__.py

@@ -1,6 +1,6 @@
 """Public exports for the selectools package."""
 
-__version__ = "0.25.1"
+__version__ = "0.27.0"
 
 # Import submodules (lazy loading for optional dependencies)
 from . import embeddings, evals, guardrails, models, observe, patterns, rag, toolbox
@@ -45,6 +45,7 @@ from .guardrails import (
     GuardrailsPipeline,
     LengthGuardrail,
     PIIGuardrail,
+    PromptInjectionGuardrail,
     TopicGuardrail,
     ToxicityGuardrail,
 )
@@ -149,7 +150,13 @@ from .pending import (
 )
 from .pipeline import Pipeline, Step, StepResult, branch, cache_step, parallel, retry, step
 from .policy import ApprovalRequest, PolicyDecision, PolicyResult, ToolPolicy
-from .pricing import PRICING, calculate_cost, calculate_embedding_cost, get_model_pricing
+from .pricing import (
+    PRICING,
+    calculate_cost,
+    calculate_cost_with_cached_input,
+    calculate_embedding_cost,
+    get_model_pricing,
+)
 from .prompt import REASONING_STRATEGIES, PromptBuilder
 from .providers.anthropic_provider import AnthropicProvider
 from .providers.azure_openai_provider import AzureOpenAIProvider
@@ -161,8 +168,20 @@ from .providers.openai_provider import OpenAIProvider
 from .providers.router import RouterConfig, RouterProvider
 from .providers.stubs import LocalProvider
 from .results import Ambiguous, Artifact, NotFound, ToolResult, emit_artifact
+from .scheduler import (
+    AgentScheduler,
+    CronSchedule,
+    IntervalSchedule,
+    JobResult,
+    Schedule,
+    ScheduledJob,
+    cron,
+    every,
+)
 from .sessions import (
+    DynamoDBSessionStore,
     JsonFileSessionStore,
+    MongoSessionStore,
     RedisSessionStore,
     SessionMetadata,
     SessionSearchResult,
@@ -246,6 +265,7 @@ __all__ = [
     # Pricing
     "PRICING",
     "calculate_cost",
+    "calculate_cost_with_cached_input",
     "calculate_embedding_cost",
     "get_model_pricing",
     # Model Registry
@@ -314,6 +334,15 @@ __all__ = [
     "RepeatDetector",
     "StallDetector",
     "PingPongDetector",
+    # Scheduled agents
+    "AgentScheduler",
+    "ScheduledJob",
+    "JobResult",
+    "Schedule",
+    "CronSchedule",
+    "IntervalSchedule",
+    "cron",
+    "every",
     # Guardrails
     "guardrails",
     "Guardrail",
@@ -326,6 +355,7 @@ __all__ = [
     "PIIGuardrail",
     "TopicGuardrail",
     "ToxicityGuardrail",
+    "PromptInjectionGuardrail",
     # Audit
     "AuditLogger",
     "PrivacyLevel",
@@ -339,6 +369,8 @@ __all__ = [
     "SQLiteSessionStore",
     "RedisSessionStore",
     "SupabaseSessionStore",
+    "MongoSessionStore",
+    "DynamoDBSessionStore",
     # Entity Memory
     "Entity",
     "EntityMemory",

{selectools-0.25.1 → selectools-0.27.0}/src/selectools/a2a/server.py

@@ -94,6 +94,10 @@ def _extract_prompt(parts: List[Dict[str, Any]]) -> str:
     """
     chunks: List[str] = []
     for part in parts:
+        if not isinstance(part, dict):
+            # Defensive: validation rejects non-dict parts up front, but a
+            # stray list element must never crash prompt assembly.
+            continue
         kind = part.get("kind")
         if kind == "text" and isinstance(part.get("text"), str):
             chunks.append(part["text"])
@@ -300,6 +304,8 @@ class A2AServer:
         parts = message.get("parts")
         if not isinstance(parts, list) or not parts:
             return None, "'message.parts' must be a non-empty array"
+        if not all(isinstance(p, dict) for p in parts):
+            return None, "every entry in 'message.parts' must be an object"
         if not any(
             isinstance(p, dict) and p.get("kind") == "text" and isinstance(p.get("text"), str)
             for p in parts

{selectools-0.25.1 → selectools-0.27.0}/src/selectools/agent/_memory_manager.py

@@ -10,6 +10,7 @@ from ..types import Message, Role
 
 if TYPE_CHECKING:
     from ..memory import ConversationMemory
+    from .core import _RunContext
 
 
 _MAX_SUMMARY_CHARS = 4000  # ~1000 tokens, well under any provider's context window
@@ -132,6 +133,21 @@ class _MemoryManagerMixin:
         except Exception:  # nosec B110
             pass  # never crash the agent for a persistence failure
 
+    def _unified_memory_record(self, ctx: "_RunContext", final_response: Message) -> None:
+        """Persist the completed turn into unified memory, if configured.
+
+        Drives ``UnifiedMemory.add_turn()`` — the single write path that
+        records the episode, feeds the entity tier (when present), and
+        promotes aged-out short-term items to long-term memory.
+        """
+        unified = getattr(self, "unified_memory", None)
+        if unified is None:
+            return
+        try:
+            unified.add_turn(ctx.unified_user_text, final_response)
+        except Exception:  # nosec B110
+            pass  # never crash the agent for a memory persistence failure
+
     def _extract_entities(self, run_id: str) -> None:
         """Extract entities from recent messages if entity_memory is configured."""
         em = self.config.entity_memory

{selectools-0.25.1 → selectools-0.27.0}/src/selectools/agent/config_groups.py

@@ -39,6 +39,7 @@ if TYPE_CHECKING:
     from ..policy import ToolPolicy
     from ..providers.base import Provider
     from ..sessions import SessionStore
+    from ..unified_memory import UnifiedMemory
     from ..usage import AgentUsage
 
 
@@ -199,11 +200,95 @@ class SummarizeConfig:
 @stable
 @dataclass
 class MemoryConfig:
-    """Memory subsystem settings."""
+    """Memory subsystem settings.
+
+    The ``unified*`` fields below are **beta** (v1.1): they wire
+    :class:`~selectools.unified_memory.UnifiedMemory` into the agent so it
+    builds context from all tiers before each call and persists the completed
+    turn (with STM -> LTM auto-promotion) after each run.
+
+    Attributes:
+        entity_memory: Optional EntityMemory for per-turn entity extraction
+            and context injection.
+        knowledge_graph: Optional KnowledgeGraphMemory for relationship
+            triple extraction and query-scoped context.
+        knowledge_memory: Optional KnowledgeMemory for durable facts and the
+            auto-injected ``remember`` tool.
+        unified: Enable unified tiered memory (beta). The agent builds a
+            :class:`~selectools.unified_memory.UnifiedMemory` from the tier
+            parameters below and drives it for context assembly and post-turn
+            persistence. Mutually exclusive with ``entity_memory``,
+            ``knowledge_graph``, ``knowledge_memory``, the Agent ``memory=``
+            parameter, and ``session_store``. Default: False.
+        unified_memory: Optional pre-built UnifiedMemory instance (beta).
+            Implies ``unified=True``. Use this to inject custom tiers
+            (``UnifiedMemory(short_term=..., long_term=..., entity_memory=...)``,
+            custom scorers, summarizers). When set, the scalar tier
+            parameters below are ignored — the instance's own settings win.
+            Default: None.
+        importance_threshold: Minimum importance score (0.0-1.0) for STM ->
+            LTM promotion. Default: 0.7.
+        short_term_limit: Rolling short-term window size, in messages
+            (one turn = two messages). Default: 100.
+        long_term_limit: Maximum long-term entries before importance-based
+            eviction. Default: 1000.
+        episodic_retention_days: Episodes older than this are pruned.
+            Default: 30.
+        auto_promote: Score and promote short-term items as they age out of
+            the rolling window. Default: True.
+        context_max_tokens: Token budget passed to
+            ``UnifiedMemory.assemble_context()`` when the agent injects
+            unified context; compaction triggers at 70% of this budget.
+            Default: 4000.
+
+    Raises:
+        ValueError: If unified memory is enabled together with
+            ``entity_memory``, ``knowledge_graph``, or ``knowledge_memory``;
+            or if any unified tier parameter is out of range while unified
+            memory is enabled (the parameters are inert otherwise).
+    """
 
     entity_memory: Optional["EntityMemory"] = None
     knowledge_graph: Optional["KnowledgeGraphMemory"] = None
     knowledge_memory: Optional["KnowledgeMemory"] = None
+    # -- Unified tiered memory (beta, v1.1) --
+    unified: bool = False
+    unified_memory: Optional["UnifiedMemory"] = None
+    importance_threshold: float = 0.7
+    short_term_limit: int = 100
+    long_term_limit: int = 1000
+    episodic_retention_days: int = 30
+    auto_promote: bool = True
+    context_max_tokens: int = 4000
+
+    @property
+    def _unified_enabled(self) -> bool:
+        """Whether unified memory is requested (flag or injected instance)."""
+        return self.unified or self.unified_memory is not None
+
+    def __post_init__(self) -> None:
+        if not self._unified_enabled:
+            return
+        if (
+            self.entity_memory is not None
+            or self.knowledge_graph is not None
+            or self.knowledge_memory is not None
+        ):
+            raise ValueError(
+                "MemoryConfig: unified memory is mutually exclusive with "
+                "entity_memory/knowledge_graph/knowledge_memory. Inject custom "
+                "tiers via MemoryConfig(unified_memory=UnifiedMemory(...)) instead."
+            )
+        if not 0.0 <= self.importance_threshold <= 1.0:
+            raise ValueError("MemoryConfig.importance_threshold must be between 0.0 and 1.0")
+        if self.short_term_limit < 1:
+            raise ValueError("MemoryConfig.short_term_limit must be at least 1")
+        if self.long_term_limit < 1:
+            raise ValueError("MemoryConfig.long_term_limit must be at least 1")
+        if self.episodic_retention_days < 1:
+            raise ValueError("MemoryConfig.episodic_retention_days must be at least 1")
+        if self.context_max_tokens < 1:
+            raise ValueError("MemoryConfig.context_max_tokens must be at least 1")
 
 
 @stable