Subagent orchestration
Most agents can spawn subagents but rarely do. AFK makes them the default. One interrupt kills the whole tree.
Start a run in your terminal and walk away. Get pinged when it finishes, or needs you. Every step is a readable trace you check before anything ships.
$npm install -g agent-afkRequires Node ≥20 · puts afk on your PATH
Free forever · No account required · Apache-2.0
53k downloads last month
Left alone, most agents chase one hypothesis, stop too early, hand you a vague done, and drift when they parallelize.
Agent AFK governs exactly these failure modes:
One hypothesis, one file, one pass: commits to the first idea, never branches.
Fixed by: Parallel hypotheses via /diagnoseClaims completion without verification: confident, unchecked, sometimes just wrong.
Fixed by: Adversarial verification via /shadow-verifyLong runs decay into unreadable soup. The longer it works, the less anyone can follow.
Fixed by: Local trace history under ~/.afk/Parallel lanes diverge with nothing aligning them. Work that should converge quietly splits.
Fixed by: One interrupt kills the whole subagent treeYou step back in and can’t tell what happened: no receipt, just scrollback.
Fixed by: Explicit terminal states + on-disk tracesYou still supervise every few minutes. “Autonomous” that needs watching isn’t.
Fixed by: Remote supervision + reversibility gatesI built this becausefinisheddoes not meancorrect. One run: an internal check approved the work, but the next stage wanted that approval in a slightly different format. Good work, rejected three times until max retries. Final answer: failure. Nothing crashed; just a tiny handoff bug dressed up as a careful decision. That is real agent failure: bad handoffs, missing context, fake done. Most products assume someone is watching. Agent AFK assumes you are not.
Agent AFK is built to change that: parallel hypotheses, isolated worktrees, and verification by default, all governed by traces, terminal states, and reversibility rules.
Orchestration, routing, traces, verification, supervision: wired into the loop, not bolted on.
Most agents can spawn subagents but rarely do. AFK makes them the default. One interrupt kills the whole tree.
Type /diagnose for a real pipeline: parallel hypotheses, isolated worktrees, verification. One command, not one long prompt and hope.
When a midnight run goes sideways, read exactly what happened: every session, subagent, and tool call on disk under ~/.afk/.
Every tool call inline: arguments, output, timing. When the agent says done, check what it actually did.
A second adversarial agent re-derives claims before they reach your diff: finished has to prove it is correct.
Leave the desk. Your phone buzzes when the run finishes or stalls: answer, redirect, or interrupt from Telegram.
The hard part is multi-step work without doing something irreversible too early. AFK builds that into the loop.
Reversible edits, isolated worktrees, scratch files: move without ceremony.
Commits, pushes, deletes, external calls: gated or verified.
Done, Blocked, Asking, Interrupted. The agent knows when to stop.
Plans, traces, and session state persist under ~/.afk/. Inspect, resume, roll back.
Tell it once. The next session reads it: preferences, conventions, and decisions land in ~/.afk/, searchable from any run.
~/.afk/One command, a full parallel pipeline: each skill fans out subagents, gathers results, and re-checks. That's the orchestration you'd otherwise ask for every time. Eleven ship in the box; /forge makes more.
End-to-end feature pipeline: spec → research → plan → parallelize → build → verify → heal → ship.
Ship-ready in one pass; pauses for approval.
Parallel hypotheses for bugs and failing tests.
Each validated in an isolated git worktree.
Adversarial re-derivation of sub-agent claims before you act.
Re-checked before the build pipeline.
Generate new skills autonomously, gated by L1 capability evals.
Finds gaps, drafts, runs a qualify loop.
Turn a linear plan into dependency-aware parallel waves.
Respects file and import dependencies.
Parallel reviewers across a diff or PR: security, correctness, api-compat, tests, perf.
Synthesized by severity into a merge recommendation.
Release pipeline for finished local work: pre-flight, tests, commit, push, PR.
With optional adversarial verification.
Turn a loose idea into a structured, actionable spec.
Phase 1 of /mint.
Two sub-agents (web context + local repo context) merged into one brief.
External and local, combined.
Three parallel critics invent alternatives; a synthesis step ranks all four.
Pragmatist, paranoid, architect lenses; a ranker picks.
Parallel context-gathering for code you're about to edit.
Structure and test-coverage agents, one wave.
All three converge on ~/.afk/: config, plugins, plans, skills, session traces, memory. Nothing leaks into your global config.
Interactive TUI. Streaming chat, plan mode, slash skills, background tasks.
Background loop. Runs unattended with bypass-permissions and plan persistence.
Mobile attach. Send messages, tail output, interrupt from anywhere.
Config · plugins · plans · skills · session traces · memory
The runtime is Apache-2.0 and free: every feature, no unlocks. You pay only for setup help and (soon) team governance.
The whole runtime: yours to run, read, fork, and ship. Nothing held back.
~/.afk/ traces & memoryI wire Agent AFK into your repo, daemon, and Telegram: with you, not just docs.
Set up directly with Griffin, the maintainer.
Typically replies within 24h.
A deeper pass on your orchestration, verification, and custom skills.
Led by Griffin Long, founder of Agent AFK.
Typically replies within 24h.
Apache-2.0 and free. Install now, or drop your email for release notes and what's next.
$npm install -g agent-afkRequires Node ≥20 · puts afk on your PATH