Skip to content

Release v0.14.0 #82

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
emal-avala merged 1 commit into from
Apr 6, 2026
Merged

Release v0.14.0 #82

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
22 changes: 20 additions & 2 deletions CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,10 +7,28 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0

## [Unreleased]

## [0.14.0] - 2026-04-06

### Added

- **`/powerup` interactive tutorials**: 5 step-by-step lessons teaching core features (first conversation, editing files, shell & tools, skills & workflows, models & providers). Arrow-key lesson picker with persistent progress tracking. Aliases: `/tutorial`, `/learn`. Reset with `/powerup reset`.
- **Homebrew tap auto-update**: release workflow now automatically updates `avala-ai/homebrew-tap` formula with new version and SHA256 checksums on every tagged release
- **Cross-platform Flutter client** with WebSocket JSON-RPC backend: desktop app (macOS, Linux), web WASM build, bidirectional permission prompting, heartbeat-based dead client detection, auto-update checker. Shared Dart client library at `packages/agent_code_client/`
- **Behavioral evaluation framework**: `crates/eval/` with `eval_test!` macro, `TestRig` with workspace setup and tool log capture, best-of-4 retry logic for LLM non-determinism, two policy tiers (`AlwaysPasses` / `UsuallyPasses`), breakpoint mechanism for model steering tests. Ships with 10 seed evals
- **Shell passthrough context injection**: `!` prefix now streams output in real-time (piped subprocess instead of `.output()`) and injects captured output into conversation history as `is_meta` message. Agent can reference shell output in subsequent turns without copy-pasting. 50KB truncation prevents context bloat
- **Prompt caching for tool definitions**: `cache_control: { type: "ephemeral" }` on the last tool in the tools array, caching the full prefix (system prompt + 32 tools). New `features.prompt_caching` config toggle (default: true)
- **Scheduled agents**: `agent cron add` for recurring agent execution, `agent trigger --listen` for webhook-triggered runs, background daemon for unattended operation
- **`/powerup` interactive tutorials**: 5 step-by-step lessons teaching core features. Arrow-key lesson picker with persistent progress tracking. Aliases: `/tutorial`, `/learn`
- **Homebrew tap auto-update**: release workflow automatically updates `avala-ai/homebrew-tap` formula with new version and SHA256 checksums on every tagged release
- **Shell passthrough test suite**: 40 tests across 3 layers (25 unit, 11 integration, 4 E2E bash). Extracted capture logic into `services/shell_passthrough.rs` for testability
- **Detailed v1.1/v1.2 roadmap**: 15 feature specifications with architecture diagrams, config schemas, and implementation checklists

### Changed

- **Shell passthrough refactored**: `!` handler in `repl.rs` reduced from 92 lines to 19 lines by extracting into `services/shell_passthrough.rs` library module
- **Prompt caching wired to config**: `enable_caching` in query loop now reads `features.prompt_caching` instead of hardcoded `true`

### Fixed

- **E2E D10 coding task**: made resilient to context overflow during long agent sessions

## [0.13.1] - 2026-04-05

Expand Down
4 changes: 2 additions & 2 deletions Cargo.lock
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

4 changes: 2 additions & 2 deletions crates/cli/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
[package]
name = "agent-code"
version = "0.13.1"
version = "0.14.0"
edition = "2024"
description = "An AI-powered coding agent for the terminal, written in pure Rust"
license = "MIT"
Expand All @@ -16,7 +16,7 @@ name = "agent"
path = "src/main.rs"

[dependencies]
agent-code-lib = { path = "../lib", version = "0.13.1" }
agent-code-lib = { path = "../lib", version = "0.14.0" }

# CLI
clap = { version = "4", features = ["derive", "env"] }
Expand Down
2 changes: 1 addition & 1 deletion crates/eval/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,7 +10,7 @@ name = "eval_runner"
path = "src/main.rs"

[dependencies]
agent-code-lib = { path = "../lib", version = "0.13.1" }
agent-code-lib = { path = "../lib", version = "0.14.0" }

# Async runtime
tokio = { version = "1", features = ["full"] }
Expand Down
2 changes: 1 addition & 1 deletion crates/lib/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
[package]
name = "agent-code-lib"
version = "0.13.1"
version = "0.14.0"
edition = "2024"
description = "Agent engine library: LLM providers, tools, query loop, memory"
license = "MIT"
Expand Down
2 changes: 1 addition & 1 deletion npm/package.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
{
"name": "@avala-ai/agent-code",
"version": "0.13.1",
"version": "0.14.0",
"description": "AI coding agent for the terminal. Built in Rust.",
"license": "MIT",
"repository": {
Expand Down
Loading