Design or run a long-lived allowlisted worker plus watchdog for multi-hour agentic work without human babysitting. Use when the user runs /autonomous-worker-ops, asks for marathon, watchdog, scheduler, 24h/48h unattended jobs, launchd/cron worker, heartbeat, pause file, or keep agents working for N hours. Does not place trades, hold secrets, or replace dual-agent review. Compose with dual-agent-review for high-blast-radius promotion.
EnglishRun a fail-closed producer/reviewer protocol for plans, diffs, claims, or designs. Use when the user runs /dual-agent-review, asks for dual-Grok, second-pair-of-eyes, hostile review, ACCEPT/REVISE/HUMANREQUIRED, handoff protocol, or independent QA before merge/freeze/spend. Portable across Grok, Claude, Codex, and other agents.
EnglishApply a fail-closed promotion ladder to any claim or feature: Forbidden, Explore, Confirm, Act (human-gated). Use when the user runs /fail-closed-promotion, asks “may we act”, promotion ladder, explore vs confirm, ship readiness, agent permission, or PATHC-style stop. Default is no production action.
EnglishGrade investment claims for stocks, ETFs, crypto, and macro using a fail-closed verdict ladder. Use when the user asks whether to buy/sell/hold, wants a second opinion on a thesis, runs /investment-claim-court, mentions claim court, Reject, Watch, Research More, Small Position Allowed, Confirmed Thesis, or Personal Investment Evidence Lab. Never give tips or price predictions; only claim grades.
English
AGENT SKILLS ATLASIncludes Autonomous Worker Ops, Dual Agent Review, Fail Closed Promotion, and other workflows.
Claude Code
Codex
Cursor
GitHub Copilot
Gemini CLISupported by the Atlas install prompt. Skill-specific requirements vary.
Pack stats
What’s included
Autonomous Worker Ops
Design or run a long-lived allowlisted worker plus watchdog for multi-hour agentic work without human babysitting. Use when the user runs /autonomous-worker-ops, asks for marathon, watchdog, scheduler, 24h/48h unattended jobs, launchd/cron worker, heartbeat, pause file, or keep agents working for N hours. Does not place trades, hold secrets, or replace dual-agent review. Compose with dual-agent-review for high-blast-radius promotion.
Dual Agent Review
Run a fail-closed producer/reviewer protocol for plans, diffs, claims, or designs. Use when the user runs /dual-agent-review, asks for dual-Grok, second-pair-of-eyes, hostile review, ACCEPT/REVISE/HUMANREQUIRED, handoff protocol, or independent QA before merge/freeze/spend. Portable across Grok, Claude, Codex, and other agents.
Fail Closed Promotion
Apply a fail-closed promotion ladder to any claim or feature: Forbidden, Explore, Confirm, Act (human-gated). Use when the user runs /fail-closed-promotion, asks “may we act”, promotion ladder, explore vs confirm, ship readiness, agent permission, or PATHC-style stop. Default is no production action.
Investment Claim Court
Grade investment claims for stocks, ETFs, crypto, and macro using a fail-closed verdict ladder. Use when the user asks whether to buy/sell/hold, wants a second opinion on a thesis, runs /investment-claim-court, mentions claim court, Reject, Watch, Research More, Small Position Allowed, Confirmed Thesis, or Personal Investment Evidence Lab. Never give tips or price predictions; only claim grades.
Quick start
- Choose your skills. Select individual skills above, or choose the entire pack.
- Copy the install prompt. Use Copy install prompt above the list.
- Send it to your agent. Paste into your coding agent’s chat. Your agent reviews the source and guides the setup.
Try it with your agent
After installation, start with a request like:
Use the Autonomous Worker Ops skill to help with my task.
Ask me for the context you need before getting started.Requirements & license
- A coding agent that can access your project files and install skills.
- Your agent checks each skill’s instructions, dependencies, and permissions during setup.
- Reported license for the skills shown: MIT. The author’s terms apply.