Evidence-driven task discipline for hard tasks. Use PROACTIVELY, mid-task as well as at the start, and specifically whenever any of these is true - the task has 3+ dependent steps; you are about to build on a file, API response, or dataset you have not opened; you are on the second failed attempt at the same fix; you are several actions into a plan with no check against intermediate results; you are about to call something done on the strength of "it ran" rather than an observed outcome; a result came back surprisingly clean; or you cannot say in one sentence what done looks like. Also use when a task keeps failing or stalling, or when the user says "five gates", "five-gate method", "fable mode", "think like Fable", "slow down and do this right", or "think this through first". Loads a five-gate task loop (scope, evidence, adversarial reasoning, verification, calibrated reporting) plus standing habits. Model-agnostic - runs on any capable LLM.
Install with
Claude CodeCodexCursorGitHub CopilotGemini CLI
Supported by the Atlas install prompt. Skill-specific requirements vary.
Skill stats
Skills1CategoryGeneralStars0
About this skill
Evidence-driven task discipline for hard tasks. Use PROACTIVELY, mid-task as well as at the start, and specifically whenever any of these is true - the task has 3+ dependent steps; you are about to build on a file, API response, or dataset you have not opened; you are on the second failed attempt at the same fix; you are several actions into a plan with no check against intermediate results; you are about to call something done on the strength of "it ran" rather than an observed outcome; a result came back surprisingly clean; or you cannot say in one sentence what done looks like. Also use when a task keeps failing or stalling, or when the user says "five gates", "five-gate method", "fable mode", "think like Fable", "slow down and do this right", or "think this through first". Loads a five-gate task loop (scope, evidence, adversarial reasoning, verification, calibrated reporting) plus standing habits. Model-agnostic - runs on any capable LLM.
Quick start
Copy the install prompt. Use the green button above this README.
Send it to your agent. Paste it into your coding agent’s chat and send.
Review the setup. Your agent checks the skill’s instructions, requirements, and installation scope.
Evidence-driven task discipline for hard tasks. Use PROACTIVELY, mid-task as well as at the start, and specifically whenever any of these is true - the task has 3+ dependent steps; you are about to build on a file, API response, or dataset you have not opened; you are on the second failed attempt at the same fix; you are several actions into a plan with no check against intermediate results; you are about to call something done on the strength of "it ran" rather than an observed outcome; a result came back surprisingly clean; or you cannot say in one sentence what done looks like. Also use when a task keeps failing or stalling, or when the user says "five gates", "five-gate method", "fable mode", "think like Fable", "slow down and do this right", or "think this through first". Loads a five-gate task loop (scope, evidence, adversarial reasoning, verification, calibrated reporting) plus standing habits. Model-agnostic - runs on any capable LLM.
About this listing
Category
General
Reported license
MIT
Listing language (estimated)
English
Review status
Not individually reviewed
Agent compatibility
Atlas installation supports Claude Code, Codex, and Cursor. Skill-specific dependencies and features are checked during setup.
Claude Code
Codex
Cursor
GitHub Copilot
Gemini CLI