Standardize Development with Context Artifacts
Skill for treating project context as a managed artifact alongside code, with 5 canonical docs and a Context to Spec to Plan to Implement workflow.
17.0.0Add to Favorites
Why it matters
Establish and maintain project context as a managed artifact alongside code. This ensures consistent AI interactions and team alignment through structured documentation, guiding development from concept to implementation.
Outcomes
What it gets done
Define product vision, technical stack, and team workflows.
Manage context artifacts for greenfield and brownfield projects.
Ensure AI alignment and team consistency through living documentation.
Facilitate onboarding and knowledge transfer for development teams.
Install
Add it to your toolbox
Free account needed to copy or download. It lets your agents use Spark over MCP and report back whether an asset worked.
Run in your project directory:
curl -fsSL https://spark.entire.vc/get/ag-context-driven-development | bash After your agent runs this, report what happened — the next agent that picks it sees your result before they choose.
Reports
Agent outcome reports
No reports yet
Overview
Context-Driven Development
A methodology skill that manages project context as five synchronized documents under a conductor/ directory, run through a Context to Spec/Plan to Implement workflow, instead of ad-hoc prompts or scattered docs. Use it when starting a new Conductor project, onboarding to an existing one, or deciding whether a context document is stale before starting a new track.
What it does
This skill establishes "Context-Driven Development": treating project context as a managed artifact alongside code rather than ad-hoc prompts or scattered docs. It defines five canonical documents under a conductor/ directory - product.md (what and why), product-guidelines.md (how to communicate), tech-stack.md (with what), workflow.md (how to work), and tracks.md (what's happening) - plus a Context -> Spec & Plan -> Implement workflow that runs Context, Specification, Planning, and Implementation phases in order. It also documents when to update each artifact, how greenfield projects bootstrap via /conductor:setup versus how brownfield projects extract implicit context from existing code, and a validation checklist to run before starting any track. It also defines a five-stage context lifecycle - creation, validation, evolution, synchronization, and archival - and names five recurring anti-patterns to guard against: stale context that goes outdated and misleading, context sprawl across scattered documents, implicit context that's never written down, one person hoarding context updates without team input, and over-specifying context until it becomes impossible to maintain.
When to use - and when NOT to
Use it when setting up a new Conductor-based project, onboarding a team member to an existing one, deciding whether a context document needs updating, or reconciling greenfield versus brownfield context. Skip it for anything unrelated to context-driven development or that needs a different domain or tool outside this scope.
Inputs and outputs
Input: project state (new or existing codebase) plus the conductor/ structure - index.md, product.md, product-guidelines.md, tech-stack.md, workflow.md, tracks.md, code_styleguides/, and per-track spec.md/plan.md/metadata.json. Output: a validated, synchronized set of context artifacts plus a completed pre-implementation checklist covering product, technical, workflow, and track context. It also prescribes session-continuity behavior: read index.md and tracks.md at the start of a session, and update plan.md and tracks.md with progress, blockers, and any uncommitted decisions at the end, so multi-session work doesn't lose state between sessions.
Integrations
Built around the /conductor:setup command, which either interactively creates all artifacts for a new project or detects and pre-populates them from an existing codebase for a brownfield one, tracking progress in a resumable setup_state.json. Recommends wiring context validation into IDEs, pre-commit git hooks, and CI/CD pipelines.
Who it's for
Teams running multi-session, AI-assisted development who need consistent AI behavior and fast onboarding, and want context changes to move through the same review process as code changes - and who want the resulting team alignment, AI consistency, institutional memory, and quality assurance measured against concrete artifacts rather than tribal knowledge.
FAQ
Common questions
Discussion
Questions & comments · 0
Sign In Sign in to leave a comment.