# Build Projects with Managed Development Context

Scattered project knowledge creates inconsistent AI output and slows team decisions. This skill organizes product, technical, workflow, and delivery context into maintained project artifacts.

## Install

```bash
npx skillstore add sickn33/context-driven-development
```

## Metadata

- Status: approved
- Slug: sickn33-context-driven-development
- Skillstore revision: r2
- Version status: missing
- Tree hash: a21da15e3d9a070a837b2024761646a5a35fbdd2f7a88d96721c902a5220ae06
- Author: sickn33
- GitHub username: sickn33
- License: MIT
- Repository: https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/context-driven-development
- Ref: f9e2c34b4f19c7f3e6b0a1e93227b5f77cc12526
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands
- Quality score: 78
- Quality tier: bronze
- Public page: https://skillstore.pages.dev/skills/sickn33-context-driven-development
- Manifest: https://skillstore.pages.dev/api/skills/sickn33-context-driven-development/manifest

## Capabilities

- Defines roles and update triggers for product, product guidelines, technology stack, workflow, and track documents.
- Guides greenfield and brownfield setup through a structured context, specification, planning, and implementation workflow.
- Provides a validation checklist for product, technical, workflow, and active-track context.
- Explains how to synchronize project documents when dependencies, features, practices, or track status change.
- Supports session continuity with concrete starting, ending, and interruption procedures.

## Use Cases

- Establish a new project context: Create a shared foundation for product goals, technology choices, team practices, and planned work before implementation begins.
- Document an existing codebase: Extract current patterns into structured artifacts, then review differences between actual and desired practices.
- Improve team continuity: Maintain track status, decisions, and session handoffs so contributors can resume work with consistent context.

## Prompt Templates

### Create core context

```
Help me define product.md, tech-stack.md, and workflow.md for [project]. Ask for missing goals, constraints, users, technologies, and quality requirements.
```

### Audit context freshness

```
Review [context documents] against [recent changes]. Identify stale statements, conflicts, missing decisions, and specific updates required before implementation.
```

### Plan a feature track

```
Using [product context], [technical context], and [workflow], draft a specification and phased plan for [feature]. Include acceptance criteria, dependencies, and validation.
```

### Reconcile brownfield context

```
Analyze [repository observations] and [existing context]. Separate confirmed patterns from assumptions, identify contradictions, and propose prioritized artifact updates with stakeholder questions.
```

## Limitations

- Assumes a Conductor-style artifact structure and slash commands that may not exist in every environment.
- Provides guidance only; it does not create files, inspect repositories, or execute validation by itself.
- Requires stakeholders to confirm product goals, standards, and whether existing context remains accurate.
- Does not replace project-specific testing, security review, or expert architectural judgment.

## Best Practices

- Read current context artifacts before drafting specifications or implementation plans.
- Update related documents together when product scope, dependencies, workflow rules, or track status changes.
- Confirm extracted brownfield assumptions with stakeholders before treating them as canonical context.

## Anti Patterns

- Do not scatter the same decision across multiple documents without a canonical source.
- Do not let generated context replace repository evidence or stakeholder confirmation.
- Do not expand context with details that do not influence decisions, behavior, or validation.

## Security Audit

- Audited at: 2026-07-23T21:06:16.672\+00:00
- Summary: All six external-command findings are Markdown formatting or fixed Conductor slash-command references, with no shell, dynamic input, or executable code. The reconnaissance finding is ordinary anti-pattern guidance; no prompt injection, exfiltration intent, or other semantic security issue was found.

## Stats

- Views: 119
- Downloads: 12
- Favorites: 0
- Popularity score: 0