GitHub Copilot Instructions — ContextOS

<!-- CONTEXTOS:START -->
<!--
Generated by ContextOS
Adapter: copilot@2
Source graph: sha256:1b5d0e2b86f8dc023ed48da798410c82c6d2855db15d6af500b575cb60234bd6
Profile: default sha256:0000000000000000000000000000000000000000000000000000000000000000
Do not edit; use .agents/project overrides.
-->
# GitHub Copilot Instructions — ContextOS

## Project Rules

### Step 0 — Identify Before Acting

When useful for substantial work, identify:
```
1. WHAT DOMAIN?   → Frontend / Backend / Architecture / Full-Stack / DevOps
2. WHAT PHASE?    → Define / Plan / Build / Verify / Review / Ship
3. WHAT ROLE?     → Declare specialist role for this phase
```

> **Anti-Spam Invariant**: Declare this strictly once at the start of a task or phase. Never repeat before intermediate tool calls or step updates.

### Non-Negotiable Rules

- **Zero-Assumption Investigation**: Never guess file paths or signatures. Inspect before modifying.
- **Zero-Placeholder Production Code**: Never emit lazy stubs, `// TODO`, or partial code.
- **Mandatory Proof-of-Work Verification**: Always run tests and validators before claiming completion.
- **Surgical Blast Radius**: Modify only files strictly within scope.
- **Fast-Track Exception**: Routine maintenance, git commands, version bumps, typo fixes, and diagnostics skip spec/plan ceremonies.

---

## Skills Library

ContextOS uses on-demand progressive disclosure. Read the relevant `SKILL.md` file only when your task matches the trigger.

### Skill Routing Table

| Trigger / Domain | Skill | Path |
| --- | --- | --- |
| `context-manager` | **context-manager** | `.agents/core/skills/context-manager/SKILL.md` |
| `context-os` | **context-os** | `.agents/core/skills/context-os/SKILL.md` |
| `engineering-workflow` | **engineering-workflow** | `.agents/core/skills/engineering-workflow/SKILL.md` |
| `gemini-precision` | **gemini-precision** | `.agents/core/skills/gemini-precision/SKILL.md` |
| `gstack-roles` | **gstack-roles** | `.agents/core/skills/gstack-roles/SKILL.md` |
| `ponytail-mindset` | **ponytail-mindset** | `.agents/core/skills/ponytail-mindset/SKILL.md` |
| `security` | **security** | `.agents/core/skills/security/SKILL.md` |

---

### context-manager
> Compatibility alias for context-os context selection and budgeting.
*Source: `.agents/core/skills/context-manager/SKILL.md` (Read on demand)*

### context-os
> Deterministic context compiler and policy engine for AI coding agents.
*Source: `.agents/core/skills/context-os/SKILL.md` (Read on demand)*

### engineering-workflow
> Scope implementation work, verify behavior, and report evidence using a proportional lifecycle.
*Source: `.agents/core/skills/engineering-workflow/SKILL.md` (Read on demand)*

### gemini-precision
> Execution guidance for Gemini: inspect code, preserve scope, and report relevant verification evidence.
*Source: `.agents/core/skills/gemini-precision/SKILL.md` (Read on demand)*

### gstack-roles
> Compatibility alias for engineering-workflow with optional specialist review perspectives.
*Source: `.agents/core/skills/gstack-roles/SKILL.md` (Read on demand)*

### ponytail-mindset
> Choose a minimal maintainable implementation for substantive Build tasks while preserving safety and verification.
*Source: `.agents/core/skills/ponytail-mindset/SKILL.md` (Read on demand)*

### Application Security
> Protect authentication, authorization, sensitive data, untrusted input, external integrations, and agent tool execution.
*Source: `.agents/core/skills/security/SKILL.md` (Read on demand)*
<!-- CONTEXTOS:END -->