Skip to content

Agent AFK

v5.19.2 Bugfix

This release fixes issues for SREs watching stability and regressions.

Published 23d AI Agents & Assistants
✓ No known CVEs patched
Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

afk agent-framework agents ai anthropic autonomous-agent
+14 more
claude claude-code cli coding-agent daemon harness-engineering llm mcp model-context-protocol openai self-improving subagent telegram-bot typescript

Summary

AI summary

Fixed subagent handling to prevent hangs when tasks complete or fan out excessively.

Full changelog

What's Changed

  • fix(agent): fire SubagentStop for naturally-completing background subagents by @griffinwork40 in https://github.com/griffinwork40/agent-afk/pull/388
  • fix(agent): bound subagent fan-out in compose/DAG layers and runWave by @griffinwork40 in https://github.com/griffinwork40/agent-afk/pull/385
  • fix(subagent): bound forked-child tool-use loop to prevent parent hang by @griffinwork40 in https://github.com/griffinwork40/agent-afk/pull/394

Full Changelog: https://github.com/griffinwork40/agent-afk/compare/v5.19.1...v5.19.2

View diff on GitHub

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Share on X Share on Bluesky

Track Agent AFK

Get notified when new releases ship.

Sign up free

About Agent AFK

Coding-agent harness built for unattended, away-from-keyboard runs across four surfaces sharing one session manager: one-shot CLI, REPL, cron-friendly headless daemon, and a Telegram bot. Editable agent loop (prompts, permission gates, model routing, explicit terminal states), MCP (stdio/HTTP/SSE + OAuth), lifecycle hooks, background subagents, plan mode, cross-session memory, and an append-only trace receipt (`afk trace show`). Works with Anthropic and any OpenAI-compatible endpoint (GPT, Codex, local MLX/llama.cpp/Ollama)

All releases →

Related context

Beta — feedback welcome: [email protected]