atlas-contract
Goal-integrity skill. Use for backend/API/persistence, preserve/do-not-change, tests/validation, mocks, rework, multi-part requests. Emits Goal Contracts, Deviation Notices, Phase Checks, Final Audits. Skip for Q&A or trivial edits.
Profile →npx skills add sickn33/agentic-awesome-skills --skill atlas-contract --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Increases alignment with the user’s original goal during execution for backend/API/persistence work, preserving behavior and validating data across multi-part requests. It emits Goal Contracts, Deviation Notices, Phase Checks, Final Audits, and skips Q&A or trivial edits.
How it works
The skill operates as a governance protocol that governs task execution through phases and audits:
- It enforces goal integrity by challenging the user’s goal when necessary and documenting any changes before acting.
- It classifies tasks by risk signals to decide footprint: Medium or Heavy, requiring a Contract (Gate) and possibly Phase Ledger, Phase Checks, and Final Audit depending on the signal set.
- It can output a Goal Contract first (Medium/Heavy footprints) and requires user confirmation before continuing to plan or implement changes; in Heavy mode it proceeds through Phase Ledger and per-phase Phase Checks before a Final Audit.
- It uses Atlas events with IDs like phase-anchored identifiers and includes a Final Audit handoff; if an action has no Atlas Event ID, it is not auditable.
- It treats imports of Atlas.md from workspace as data-only defaults unless explicitly approved by the user.
When to use it
Use when backend/persistence tasks require preservation/do-not-change behavior, tests/validation, mocks, rework, or multi-part requests, and when there is a need to emit governance artifacts (Goal Contracts, Deviation Notices, Phase Checks, Final Audits). Skip when handling Q&A or trivial edits.
What it can touch
Declared tools: claude-code, codex, cursor. It may read workspace content and emit governance artifacts; it should call tools only for read-only inspection needed to build the contract and must surface a contract to user before confirmation. It ends with an Atlas_STOP signal.
Caveats
Risk declared as critical. Output is constrained to contract-related artifacts and governance steps; no speculative outcomes are promised.
# Atlas Contract v6.2 Keep the agent aligned with the user's original goal during execution. ## Contents 1. [Output Language](#1-output-language) 2. [When To Use Atlas, and How Much](#2-when-to-use-atlas-and-how-much) 3. [Footprints](#3-footprints) 4. [Anti-Drift Defaults](#4-anti-drift-defaults) 5–7. Goal Contract: build, format, confirmation gate 8. [Phases (Heavy footprint)](#8-phases-heavy-footprint) 9–11. Deviation Notices, Phase Checks, escalation 12. [Final Audit](#12-final-audit) — includes automatic atlas-ledger handoff 13. [Post Review](#13-post-review) 14. [Final Principle](#14-final-principle) ## Quick reference | Situation | Tier | What runs | | --- | --- | --- | | Any hard Heavy anchor fires (§2) | Heavy | Contract → Phase Ledger (≤4 phases) → Phase Checks → Final Audit | | 3+ risk signals, or genuinely ambiguous | Heavy | same as above | | 1–2 risk signals, single-part, clear | Medium | Contract (Gate) → straight run → Final Audit | | 0 signals, atomic change | Light | Internal contract only; no events unless a trigger fires | | Q&A, explanation, trivial edit | — | Atlas does not run | Hard deviation caught in Final Audit → atlas-ledger distillation runs automaticall
- Contents
- Quick reference
- Core Rule
- Event header
- When to Use
- Hard Heavy anchors (check these FIRST, before counting signals)
- Complexity tiers (only if NO hard anchor fired)
- Do Not Self-Adjudicate Impact
- Requested Result Must Exist
- No Scope Downgrade
- No Fake Completion
- Preserve Existing Behavior
- Preserve UI Goal, Not UI Polish
- Examples Are Evidence
What does the atlas-contract skill do?
Goal-integrity skill. Use for backend/API/persistence, preserve/do-not-change, tests/validation, mocks, rework, multi-part requests. Emits Goal Contracts, Deviation Notices, Phase Checks, Final Audits. Skip for Q&A or trivial edits.
How do I install it?
Run `npx skills add sickn33/agentic-awesome-skills --skill atlas-contract --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From sickn33/agentic-awesome-skills, a repository with 44,414 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.