End-of-task and resume-time checks for the recurring process-discipline failure modes: source-of-truth selection, wiring-vs-backend completion, deploy-validation reads signal not metadata, multi-axis brief discipline, session-resume state drift, branch/worktree hygiene, recommendation followthrough. Mechanical where possible; agent-facing checklist where not. Triggers on 'counter-patterns', 'before I declare done', 'am I drifting', 'resume check', 'pre-compaction checkpoint', 'post-compaction reconcile'. Distinct from /honesty-audit (lint-tier mechanical scan for stubs/bypasses) and /deep-review (broad multi-axis review): this skill enforces the process discipline counter-patterns. Three modes: --check (run end-of-task checks), --resume (post-compaction reconciliation), --checkpoint (write AGENTSTATE.md before risky long-running ops).
Install with
Claude CodeCodexCursorGitHub CopilotGemini CLI
Supported by the Atlas install prompt. Skill-specific requirements vary.
Skill stats
Skills1CategoryEngineeringStars0
About this skill
End-of-task and resume-time checks for the recurring process-discipline failure modes: source-of-truth selection, wiring-vs-backend completion, deploy-validation reads signal not metadata, multi-axis brief discipline, session-resume state drift, branch/worktree hygiene, recommendation followthrough. Mechanical where possible; agent-facing checklist where not. Triggers on 'counter-patterns', 'before I declare done', 'am I drifting', 'resume check', 'pre-compaction checkpoint', 'post-compaction reconcile'. Distinct from /honesty-audit (lint-tier mechanical scan for stubs/bypasses) and /deep-review (broad multi-axis review): this skill enforces the process discipline counter-patterns. Three modes: --check (run end-of-task checks), --resume (post-compaction reconciliation), --checkpoint (write AGENTSTATE.md before risky long-running ops).
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.
End-of-task and resume-time checks for the recurring process-discipline failure modes: source-of-truth selection, wiring-vs-backend completion, deploy-validation reads signal not metadata, multi-axis brief discipline, session-resume state drift, branch/worktree hygiene, recommendation followthrough. Mechanical where possible; agent-facing checklist where not. Triggers on 'counter-patterns', 'before I declare done', 'am I drifting', 'resume check', 'pre-compaction checkpoint', 'post-compaction reconcile'. Distinct from /honesty-audit (lint-tier mechanical scan for stubs/bypasses) and /deep-review (broad multi-axis review): this skill enforces the *process discipline* counter-patterns. Three modes: --check (run end-of-task checks), --resume (post-compaction reconciliation), --checkpoint (write AGENT_STATE.md before risky long-running ops).
About this listing
Category
Engineering
Reported license
BSD-3-Clause
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