release: v0.11.3

CHANGELOG.md

@@ -6,6 +6,8 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
 
 ## [Unreleased]
 
+## [0.11.3] - 2026-02-20
+
 ### Added
 - `LoopbackChannel` / `LoopbackHandle` / `LoopbackEvent` in zeph-core — headless channel for daemon mode, pairs with a handle that exposes `input_tx` / `output_rx` for programmatic agent I/O
 - `ProcessorEvent` enum in zeph-a2a server — streaming event type replacing synchronous `ProcessResult`; `TaskProcessor::process` now accepts an `mpsc::Sender<ProcessorEvent>` and returns `Result<(), A2aError>`
@@ -1051,7 +1053,8 @@ let agent = Agent::new(provider, channel, &skills_prompt, executor);
 - Agent calls channel.send_typing() before each LLM request
 - Agent::run() uses tokio::select! to race channel messages against shutdown signal
 
-[Unreleased]: https://github.com/bug-ops/zeph/compare/v0.11.2...HEAD
+[Unreleased]: https://github.com/bug-ops/zeph/compare/v0.11.3...HEAD
+[0.11.3]: https://github.com/bug-ops/zeph/compare/v0.11.2...v0.11.3
 [0.11.2]: https://github.com/bug-ops/zeph/compare/v0.11.1...v0.11.2
 [0.11.1]: https://github.com/bug-ops/zeph/compare/v0.11.0...v0.11.1
 [0.11.0]: https://github.com/bug-ops/zeph/compare/v0.10.0...v0.11.0

Cargo.toml

@@ -5,7 +5,7 @@ resolver = "3"
 [workspace.package]
 edition = "2024"
 rust-version = "1.88"
-version = "0.11.2"
+version = "0.11.3"
 authors = ["bug-ops"]
 license = "MIT"
 repository = "https://github.com/bug-ops/zeph"
@@ -89,18 +89,18 @@ unicode-width = "0.2"
 url = "2.5"
 uuid = "1.21"
 wiremock = "0.6.5"
-zeph-a2a = { path = "crates/zeph-a2a", version = "0.11.2" }
-zeph-channels = { path = "crates/zeph-channels", version = "0.11.2" }
-zeph-core = { path = "crates/zeph-core", version = "0.11.2" }
-zeph-gateway = { path = "crates/zeph-gateway", version = "0.11.2" }
-zeph-index = { path = "crates/zeph-index", version = "0.11.2" }
-zeph-llm = { path = "crates/zeph-llm", version = "0.11.2" }
-zeph-mcp = { path = "crates/zeph-mcp", version = "0.11.2" }
-zeph-memory = { path = "crates/zeph-memory", version = "0.11.2" }
-zeph-scheduler = { path = "crates/zeph-scheduler", version = "0.11.2" }
-zeph-skills = { path = "crates/zeph-skills", version = "0.11.2" }
-zeph-tools = { path = "crates/zeph-tools", version = "0.11.2" }
-zeph-tui = { path = "crates/zeph-tui", version = "0.11.2" }
+zeph-a2a = { path = "crates/zeph-a2a", version = "0.11.3" }
+zeph-channels = { path = "crates/zeph-channels", version = "0.11.3" }
+zeph-core = { path = "crates/zeph-core", version = "0.11.3" }
+zeph-gateway = { path = "crates/zeph-gateway", version = "0.11.3" }
+zeph-index = { path = "crates/zeph-index", version = "0.11.3" }
+zeph-llm = { path = "crates/zeph-llm", version = "0.11.3" }
+zeph-mcp = { path = "crates/zeph-mcp", version = "0.11.3" }
+zeph-memory = { path = "crates/zeph-memory", version = "0.11.3" }
+zeph-scheduler = { path = "crates/zeph-scheduler", version = "0.11.3" }
+zeph-skills = { path = "crates/zeph-skills", version = "0.11.3" }
+zeph-tools = { path = "crates/zeph-tools", version = "0.11.3" }
+zeph-tui = { path = "crates/zeph-tui", version = "0.11.3" }
 
 [workspace.lints.clippy]
 all = "warn"

README.md

@@ -205,7 +205,7 @@ flowchart TD
 | **File sandbox** | Writes outside allowed paths |
 | **Tool permissions** | Glob-based allow/ask/deny policy per tool |
 | **Destructive command gate** | Accidental `rm -rf`, `DROP TABLE`, etc. |
-| **Secret redaction** | API keys leaking into context or logs (6 provider patterns) |
+| **Secret redaction** | API keys leaking into context or logs (9 provider patterns, regex-based) |
 | **SSRF protection** | Agent and MCP client requests to internal networks |
 | **Audit logging** | Full tool execution trace for forensics |
 | **Rate limiter** | TTL-based eviction, per-IP limits on gateway |
@@ -262,7 +262,7 @@ Skills **evolve**: failure detection triggers self-reflection, and the agent gen
 
 | Protocol | What It Does |
 |----------|-------------|
-| **MCP** | Connect external tool servers (stdio + HTTP) with SSRF protection |
+| **MCP** | Connect external tool servers (stdio + HTTP) with SSRF protection, command allowlist, and env var blocklist |
 | **A2A** | Agent-to-agent communication via JSON-RPC 2.0 with SSE streaming |
 | **Audio input** | Speech-to-text via OpenAI Whisper API or local Candle Whisper (offline, feature-gated); Telegram and Slack audio files transcribed automatically |
 | **Vision** | Image input via CLI (`/image`), TUI (`/image`), and Telegram photo messages; supported by Claude, OpenAI, and Ollama providers (20 MB max, automatic MIME detection) |

crates/zeph-a2a/README.md

@@ -9,7 +9,7 @@ Implements the Agent-to-Agent (A2A) protocol over JSON-RPC 2.0, enabling Zeph to
 ## Key Modules
 
 - **client** — `A2aClient` for sending tasks and messages to remote agents
-- **server** — `A2aServer` exposing an A2A-compliant endpoint (requires `server` feature)
+- **server** — `A2aServer` exposing an A2A-compliant endpoint with `ProcessorEvent` streaming via `mpsc::Sender` (requires `server` feature)
 - **card** — `AgentCardBuilder` for constructing agent capability cards
 - **discovery** — `AgentRegistry` for agent lookup and registration
 - **jsonrpc** — JSON-RPC 2.0 request/response types

crates/zeph-core/README.md

@@ -10,16 +10,21 @@ Core orchestration crate for the Zeph agent. Manages the main agent loop, bootst
 
 | Module | Description |
 |--------|-------------|
-| `agent` | `Agent` — main loop driving inference and tool execution |
+| `agent` | `Agent<C>` — main loop driving inference and tool execution; ToolExecutor erased via `Box<dyn ErasedToolExecutor>` |
+| `agent::tool_execution` | Tool call handling, redaction, and result processing |
+| `agent::message_queue` | Message queue management |
+| `agent::builder` | Agent builder API |
+| `agent::commands` | Chat command dispatch (skills, feedback, etc.) |
+| `agent::utils` | Shared agent utilities |
 | `bootstrap` | `AppBuilder` — fluent builder for application startup |
-| `channel` | `Channel` trait defining I/O adapters; `Attachment` / `AttachmentKind` for multimodal inputs (images, audio) |
-| `config` | TOML config with `ZEPH_*` env overrides |
+| `channel` | `Channel` trait defining I/O adapters; `LoopbackChannel` / `LoopbackHandle` for headless daemon I/O; `Attachment` / `AttachmentKind` for multimodal inputs |
+| `config` | TOML config with `ZEPH_*` env overrides; typed `ConfigError` (Io, Parse, Validation, Vault) |
 | `context` | LLM context assembly from history, skills, memory |
 | `cost` | Token cost tracking and budgeting |
-| `daemon` | Background daemon mode (optional feature) |
+| `daemon` | Background daemon mode with PID file lifecycle (optional feature) |
 | `metrics` | Runtime metrics collection |
 | `project` | Project-level context detection |
-| `redact` | Sensitive data redaction |
+| `redact` | Regex-based secret redaction (AWS, OpenAI, Anthropic, Google, GitLab, HuggingFace, npm, Docker) |
 | `vault` | Secret storage and resolution via vault providers (age-encrypted read/write) |
 | `diff` | Diff rendering utilities |
 | `pipeline` | Composable, type-safe step chains for multi-stage workflows |

crates/zeph-llm/README.md

@@ -10,18 +10,20 @@ Defines the `LlmProvider` trait and ships concrete backends for Ollama, Claude,
 
 | Module | Description |
 |--------|-------------|
-| `provider` | `LlmProvider` trait — unified inference interface |
+| `provider` | `LlmProvider` trait — unified inference interface; `name()` returns `&str` (no longer `&'static str`) |
 | `ollama` | Ollama HTTP backend |
-| `claude` | Anthropic Claude backend |
-| `openai` | OpenAI backend |
+| `claude` | Anthropic Claude backend with `with_client()` builder for shared `reqwest::Client` |
+| `openai` | OpenAI backend with `with_client()` builder for shared `reqwest::Client` |
 | `compatible` | Generic OpenAI-compatible endpoint backend |
 | `candle_provider` | Local inference via Candle (optional feature) |
-| `orchestrator` | Multi-model coordination and fallback |
+| `orchestrator` | Multi-model coordination and fallback; `send_with_retry()` helper deduplicates retry logic |
 | `router` | Model selection and routing logic |
 | `vision` | Image input support — base64-encoded images in LLM requests; optional dedicated `vision_model` per provider |
-| `extractor` | `chat_typed<T>()` — typed LLM output via JSON Schema (`schemars`) |
+| `extractor` | `chat_typed<T>()` — typed LLM output via JSON Schema (`schemars`); per-`TypeId` schema caching |
+| `sse` | Shared `sse_to_chat_stream()` helpers for Claude and OpenAI SSE parsing |
 | `stt` | `SpeechToText` trait and `WhisperProvider` (OpenAI Whisper, feature-gated behind `stt`) |
 | `candle_whisper` | Local offline STT via Candle (whisper-tiny/base/small, feature-gated behind `candle`) |
+| `http` | `default_client()` — shared HTTP client with standard timeouts and user-agent |
 | `error` | `LlmError` — unified error type |
 
 **Re-exports:** `LlmProvider`, `LlmError`

crates/zeph-mcp/README.md

@@ -9,7 +9,7 @@ Implements the Model Context Protocol client for Zeph, managing connections to m
 ## Key Modules
 
 - **client** — low-level MCP transport and session handling
-- **manager** — `McpManager`, `McpTransport`, `ServerEntry` for multi-server lifecycle
+- **manager** — `McpManager`, `McpTransport`, `ServerEntry` for multi-server lifecycle; command allowlist validation (npx, uvx, node, python3, docker, etc.), env var blocklist (LD_PRELOAD, DYLD_*, NODE_OPTIONS, etc.), and path separator rejection
 - **executor** — `McpToolExecutor` bridging MCP tools into the `ToolExecutor` trait
 - **registry** — `McpToolRegistry` for tool lookup and optional Qdrant-backed search
 - **tool** — `McpTool` wrapper with schema and metadata

crates/zeph-memory/README.md

@@ -23,10 +23,11 @@ Includes a document ingestion subsystem for loading, chunking, and storing user
 | `document::pipeline` | `IngestionPipeline` — load, split, embed, store via Qdrant |
 | `vector_store` | `VectorStore` trait and `VectorPoint` types |
 | `embedding_store` | `EmbeddingStore` — high-level embedding CRUD |
+| `embeddable` | `Embeddable` trait and `EmbeddingRegistry<T>` — generic Qdrant sync/search for any embeddable type |
 | `types` | `ConversationId`, `MessageId`, shared types |
 | `error` | `MemoryError` — unified error type |
 
-**Re-exports:** `MemoryError`, `QdrantOps`, `ConversationId`, `MessageId`, `Document`, `DocumentLoader`, `TextLoader`, `TextSplitter`, `IngestionPipeline`, `Chunk`, `SplitterConfig`, `DocumentError`, `DocumentMetadata`, `PdfLoader` (behind `pdf` feature)
+**Re-exports:** `MemoryError`, `QdrantOps`, `ConversationId`, `MessageId`, `Document`, `DocumentLoader`, `TextLoader`, `TextSplitter`, `IngestionPipeline`, `Chunk`, `SplitterConfig`, `DocumentError`, `DocumentMetadata`, `PdfLoader` (behind `pdf` feature), `Embeddable`, `EmbeddingRegistry`
 
 ## Features
 

crates/zeph-scheduler/README.md

@@ -47,7 +47,7 @@ scheduler.register_handler(&TaskKind::UpdateCheck, Box::new(handler));
 Notification format sent via the channel:
 
 ```
-New version available: v0.12.0 (current: v0.11.2).
+New version available: v0.12.0 (current: v0.11.3).
 Update: https://github.com/bug-ops/zeph/releases/tag/v0.12.0
 ```
 

crates/zeph-tools/README.md

@@ -11,9 +11,9 @@ Defines the `ToolExecutor` trait for sandboxed tool invocation and ships concret
 | Module | Description |
 |--------|-------------|
 | `executor` | `ToolExecutor` trait, `ToolOutput`, `ToolCall` |
-| `shell` | Shell command executor |
+| `shell` | Shell command executor with tokenizer-based command detection, escape normalization, and transparent wrapper skipping |
 | `file` | File operation executor |
-| `scrape` | Web scraping executor |
+| `scrape` | Web scraping executor with SSRF protection (post-DNS private IP validation, pinned address client) |
 | `composite` | `CompositeExecutor` — chains executors with middleware |
 | `filter` | Output filtering pipeline |
 | `permissions` | Permission checks for tool invocation |

crates/zeph-tui/README.md

@@ -10,6 +10,7 @@ Provides a terminal UI for monitoring the Zeph agent in real time. Built on rata
 
 - **app** — `App` state machine driving the render/event loop
 - **channel** — `TuiChannel` implementing the `Channel` trait for agent I/O
+- **command_palette** — fuzzy-matching command palette with daemon commands (`daemon:connect`, `daemon:disconnect`, `daemon:status`), action commands (`app:quit`, `app:help`, `session:new`, `app:theme`), and keybinding hints
 - **event** — `AgentEvent`, `AppEvent`, `EventReader` for async event dispatch
 - **file_picker** — `@`-triggered fuzzy file search with `nucleo-matcher` and `ignore` crate
 - **highlight** — syntax highlighting for code blocks
@@ -18,6 +19,7 @@ Provides a terminal UI for monitoring the Zeph agent in real time. Built on rata
 - **metrics** — `MetricsCollector`, `MetricsSnapshot` for live telemetry
 - **theme** — color palette and style definitions
 - **widgets** — reusable ratatui widget components
+- **error** — `TuiError` typed error enum (Io, Channel)
 
 ## Usage
 

crates/zeph-tui/src/widgets/snapshots/zeph_tui__widgets__splash__tests__splash_default.snap

@@ -1,6 +1,5 @@
 ---
 source: crates/zeph-tui/src/widgets/splash.rs
-assertion_line: 76
 expression: output
 ---
 ┌──────────────────────────────────────────────────────────┐
@@ -15,7 +14,7 @@ expression: output
 │              ███████╗███████╗██║     ██║  ██║            │
 │              ╚══════╝╚══════╝╚═╝     ╚═╝  ╚═╝            │
 │                                                          │
-│                          v0.11.2                         │
+│                          v0.11.3                         │
 │                                                          │
 │                  Type a message to start.                │
 │                                                          │

docs/src/advanced/tui.md

@@ -29,7 +29,7 @@ When using `--connect`, the TUI renders token-by-token streaming from the remote
 
 ```text
 +-------------------------------------------------------------+
-| Zeph v0.11.2 | Provider: orchestrator | Model: claude-son... |
+| Zeph v0.11.3 | Provider: orchestrator | Model: claude-son... |
 +----------------------------------------+--------------------+
 |                                        | Skills (3/15)      |
 |                                        | - setup-guide      |
@@ -104,7 +104,7 @@ Press `Ctrl+P` in Insert mode to open the command palette. The palette provides
 
 | Key | Action |
 |-----|--------|
-| Any character | Filter commands by substring match |
+| Any character | Filter commands by fuzzy match |
 | `Up` / `Down` | Navigate the command list |
 | `Enter` | Execute selected command |
 | `Backspace` | Remove last query character |