Aiden AI工作流引擎

简介

<img width="1672" height="941" alt="AIDEN BOOTUP LOGO" src="https://github.com/user-attachments/assets/c0809009-73e2-4d58-9292-12fbd0324952" />

█████╗  ██╗██████╗ ███████╗███╗   ██╗
██╔══██╗██║██╔══██╗██╔════╝████╗  ██║
███████║██║██║  ██║█████╗  ██╔██╗ ██║
██╔══██║██║██║  ██║██╔══╝  ██║╚██╗██║
██║  ██║██║██████╔╝███████╗██║ ╚████║
╚═╝  ╚═╝╚═╝╚═════╝ ╚══════╝╚═╝  ╚═══╝

Autonomous Work Engine — plans, acts, recovers, and proves the result

Bundled Skills · durable tools · multiple providers · connected channels · AGPL-3.0

Windows · Linux · WSL · macOS (API Mode)

What's new in v4.19.0

Aiden v4.19.0 is the stable Worker and Model Bridge release.

• Durable Worker contracts. Read-only repository Workers use explicit parent and child execution authority.
• Immutable model binding. Each Worker keeps its provider and model binding fixed for the admitted execution.
• Accounting and budgets. Logical provider calls, physical attempts, tokens, usage, reservations, and pre-send budget enforcement remain distinct and durable.
• Resilient provider work. Bounded retries, provider fallback, cancellation during provider work, and rejection of late provider results protect current execution truth.
• Restart and evidence. Workers can restart and reopen; child Evidence is verified by the parent before completion is accepted.
• Bounded parallelism. Parallel Worker admission and reservations use stable joins and deterministic reconciliation.

What's new in v4.18.0

Aiden v4.18.0 introduces durable Codebase Mode and a substantially rebuilt Operator TUI.

What's new in v4.17.0

Aiden v4.17.0 unified durable runtime authority with a responsive operator terminal.

• One durable lifecycle. Interactive, one-shot, API, Workbench, daemon, channel, child, and MCP execution share authoritative Job and Attempt transitions.
• Stale-result protection. Generations, leases, fence tokens, terminal-state guards, and persist-first cancellation prevent old work from rewriting current truth.
• Evidence-based completion. Proof, Verdict, exact file readback, Effect reconciliation, and failure-injection coverage strengthen completion claims and recovery.
• Operator TUI foundation. The Aiden startup identity, responsive metadata, composer, status strip, themes, compact outcomes, and durable queue controls share one terminal owner.
• Platform reliability. Windows process cleanup, ConPTY coverage, and cross-platform CI preserve clean startup, activity rendering, and shutdown.

See the full v4.17.0 release notes.

What's new in v4.13

Aiden can now take a job, verify it, and keep going safely.

• Durable autonomy. Every task gets a job-card; Aiden verifies work before claiming success, resumes after restart, avoids duplicate sends/actions, retries recoverable failures, and stops for permissions.
• Evidence-required subagents. Helper agents must return real proof — a file, test result, id, or similar artifact — and Aiden re-checks that proof before trusting the result.
• Trust dial. Choose how much Aiden can do: Observer, Assistant, or Partner. Destructive actions and spending still ask, even at the highest setting.
• Glass dashboard + operator TUI. Watch model, token, budget, elapsed time, and tool steps live. Queue a follow-up, redirect mid-task, or interrupt cleanly while Aiden works.
• One-shot mode + polish. aiden -q "..." for scripting, Discord setup inside the terminal, central cross-platform path handling, and green CI across Windows, macOS, and Linux.

What's new in v4.12

Aiden can now connect outward to MCP servers and operate the browser with more state.

• MCP client. Connect to external MCP servers over stdio and Streamable HTTP, with OAuth 2.0 discovery, dynamic client registration, PKCE, token refresh, reconnects, circuit breakers, and one-command curated installs.
• Approval-gated MCP tools. External tools are model-callable, but gated by approval. Results are redacted and fenced as untrusted input.
• Browser depth. Accessibility-tree snapshots, stable element handles, act-by-reference click/type/fill, stale-element re-resolution, visual screenshot-to-vision support, multi-tab/dialog handling, and local browser attach/launch.
• Safety breadth. Cross-session search, posture-aware skill/tool narrowing, truncate-then-redact tool-output caps, per-session budget enforcement, honest execution-policy messaging, and exfiltration-aware browser guardrails.
• Process manager + activity. Tracked spawns, reliable tree-kill, redacted process listing, shutdown reaping, plus /home and /activity commands.

What's new in v4.11

A reliability release for honest output, better control, and stronger setup.

• Honesty & verification. Per-turn outcome verifier catches failed tool results and contradictions instead of glossing over them.
• Artifact registry. /artifacts tracks files Aiden creates, with provenance: source tool, originating request, and turn link.
• Streaming & control. Streaming delta coalescer reduces terminal writes, while /undo, /retry, /compress, and /usage give direct operator control.
• Task durability. Orphan-task boot sweep retires crashed-session tasks as interrupted; /tasks all shows tasks across sessions.
• Setup polish. Wizard back-navigation and config detection fixes prevent working OAuth/env setups from being mistaken as empty.
• Model updates. DeepSeek V4 Pro and V4 Flash added, model picker table improved, and weak-model UI gating reduces markup leaks.

What's new in v4.10

Aiden becomes inspectable, durable, and finally streams properly.

• Streaming default restored at the wizard layer. display.streaming now defaults to true for fresh installs — restoring what v4.1.4 attempted but couldn't operationally land because the wizard wrote the older default into your config.yaml. Existing users with explicit streaming: false keep their setting and see a one-time per-session disclosure suggesting the flip.
• Durable Task object. Every user prompt creates a persistent task row with status / goal / event back-references. /tasks [active|completed|cancelled|<N>] lists; /adjust <task_id> cancel or /adjust <task_id> goal <new text> controls.
• /trace recent event inspection. Every tool call, permission decision, and task lifecycle event lands in a queryable run_events ledger. Filter by category / kind / name / tool_call_id; scope to the current run, this session, the last N hours, or all.
• Bug D ghost-text cursor fix. Ghost suggestions now render on a footer line below the prompt; cursor lands right after your typed input. Third attempt that actually works — prior two attempts shipped inert because they tried to fight @inquirer/core's cursor manager from inside the rendered string.
• Permission picker scope clarity. Session (this path) / Always (this command) labels carry a dynamic qualifier showing what the scope actually covers.
• Eval harness v1. Contract tests under tests/v4/eval/ protect the v4.10 substrate from regression — PTY-driven where possible, mockProvider-backed for streaming guarantees.

What's new in v4.9.0

Aiden v4.9.0 ships three new feature families:

• Memory — three-namespace persistent memory with CLI and auto-review
• Hooks — secure subprocess hook system with observe, decide, and transform modes plus a full audit trail
• Strategic substrate — UUIDv7 IDs, W3C trace propagation, durable runs, idempotency, and crash recovery

Plus the theme system, MCP integration for supported clients, and substantial UI polish.

See CHANGELOG.md for details.

<br>

What's new in v4.8

Aiden's UI got an identity, and pasting actually works now.

What's new in v4.6

Aiden now spawns workers and learns from itself.

• Sub-agents. spawn_sub_agent runs a focused child with an isolated context and intersected toolset; subagent_fanout runs multiple children with merge strategies and provider rotation across configured fallback slots.
• Operator kill-switch. /spawn-pause on|off|status blocks new sub-agent spawning while in-flight children continue.
• Self-improvement loop foundation. TCE classifications and recoveries persist to durable SQLite records; /recovery list|show|clear surfaces recurring failure patterns.
• REPL parent-run lineage. Each REPL turn writes a run row; sub-agent children link back through parent lineage.
• PlannerGuard opt-in. Keyword-based per-turn tool narrowing remains optional for smaller local models.
• Onboarding redesign. Provider picker, connection probe, success screen, and /walkthrough guided tour.

<br>

What's new in v4.5

Aiden now wakes up by itself.

• Persistent daemon mode (opt-in). AIDEN_DAEMON=1 boots a background service with a SQLite-backed trigger bus.
• Autonomous trigger dispatch. File watchers, webhooks, email polling, and schedules can launch real agent turns.
• Execution guardrails + risk tiers. Filesystem boundaries, dry-run preflight, and optional Docker-backed execution.
• State-aware browser depth. URL, DOM, iframe, and interaction-state capture before and after browser actions.
• Continuous error recovery. Failure classification, bounded retries, cooldowns, dead-letter handling, and recovery reports.
• Live-flip slash commands. Toggle sandbox, TCE, browser depth, daemon behavior, suggestions, and updates without restarting.
• Update notification + /update CLI.

Full v4.5 internals: docs/v4.5/.

</details>

<br>

Core capabilities

<br>