Temp Agent Autonomy Kit

Stop waiting for prompts. Keep working.

Security Scan

The skill's instructions match its stated purpose (making agents proactive), but they are open-ended and enable continuous autonomous operation and external communication without specifying safeguards or required permissions — this raises reasonable concerns a user should review before enabling.

Purpose & Capability

Name/description (make agents proactive) align with everything in SKILL.md and README: creating task queues, replacing passive heartbeats with proactive work, cron jobs, and team channels are all coherent features for an autonomy kit. The skill does not request unrelated binaries, env vars, or installs.

Instruction Scope

Instructions are intentionally broad: 'do meaningful work', 'use remaining time', frequent heartbeats, and cron jobs give the agent wide discretion about what it does during each run. The skill references reading and writing memory files and posting to team channels (Discord/Slack) — actions that could transmit sensitive material. The SKILL.md does not define strict limits, approval gates, or exactly what kinds of data the agent may access or share.

Install Mechanism

Instruction-only skill with no install spec and no code files; nothing will be written or executed on install by the skill itself. Low install risk.

Credentials

The skill does not request environment variables or credentials itself. However, it instructs integration with team channels (Discord/Slack) and suggests using platform cron features; those integrations will require credentials outside the skill. Users should ensure any channel tokens or webhooks granted have minimal scopes and are not shared unnecessarily.

Persistence & Privilege

The skill does not set always: true and is user-invocable only. However, it explicitly recommends cron jobs and frequent autonomous heartbeats, which increase runtime persistence and blast radius. Autonomous invocation is platform-default, so the main issue is the combination of continuous schedules + open-ended work instructions rather than a metadata privilege flag.

What to consider before installing

This kit is coherent for making agents proactive, but it grants broad operational freedom. Before installing: (1) limit heartbeat frequency and token budget (e.g., longer intervals, strict run-time token caps); (2) require explicit human approval for tasks that access sensitive files or external systems; (3) scope any channel/webhook credentials tightly (create service accounts or webhooks with minimal permissions); (4) restrict which filesystem paths the agent may read/write (avoid exposing entire memory/drive); (5) audit logs and alerts for automated bursts of activity; and (6) test in an isolated session first. If you need more assurance, ask the author to add explicit safety gates (task categories that require human confirmation, token/cost safeguards, and an allowlist of accessible files and channels).

Agent Autonomy Kit

Transform your agent from reactive to proactive.

Quick Start

1. Create tasks/QUEUE.md with Ready/In Progress/Blocked/Done sections
2. Update HEARTBEAT.md to pull from queue and do work
3. Set up cron jobs for overnight work and daily reports
4. Watch work happen without prompting

Key Concepts

• Task Queue — Always have work ready
• Proactive Heartbeat — Do work, don't just check
• Continuous Operation — Work until limits hit

See README.md for full documentation.