Skip to main content
dailephd LLC
ActiveConnected product family

my-dev-kit Ecosystem

Local-first evidence and workflow infrastructure for disciplined AI-assisted software development.

A local-first toolkit that combines static repository evidence, workflow orchestration, browser/runtime evidence, and optional assurance for coding-agent development.

my-dev-kit retrieves bounded static evidence from a codebase. my-dev-kit-orchestrator organizes staged prompts, artifacts, readiness checks, responsibility continuity, and correction routing without running the coding agent itself. An external human or coding agent edits the target source. my-frontend-observer captures and evaluates what the frontend actually rendered and returns bounded runtime evidence for review or correction. my-dev-kit-lab provides optional experiments, audits, security validation, and reporting. The tools compose explicitly rather than silently invoking one another.

Designed for

  • Developers using AI-assisted coding workflows
  • Technical project maintainers
  • Teams evaluating repeatable coding-agent practices

How the products are related

Static repository evidence feeds staged workflow control; an external implementation actor changes source; Observer returns browser/runtime evidence for review and correction; Lab provides optional assurance when explicitly invoked.

my-dev-kit produces bounded static repository evidence that can be supplied to my-dev-kit-orchestrator. The orchestrator manages staged prompts, artifacts, readiness, responsibility continuity, and correction routing but does not execute the coding agent. A human or coding agent edits target source. my-frontend-observer evaluates the rendered frontend and returns runtime evidence for review or correction without editing source. my-dev-kit-lab is an optional assurance companion for supported experiments, audits, security validation, and reports.

Codebase Intelligence

my-dev-kit

Static repository evidence

Local-first indexing and graph-guided retrieval of bounded codebase evidence.

Indexes supported source structures and exposes search, exact lookup, graph slices, bounded source, role-aware context, graph diff, and conservative frontend and Android static evidence. Static evidence is not runtime proof.

Internal workflow

Index source roots
Manifest + semantic graphs
Search
Lookup
Slice
Source
Context
Bounded static evidence
bounded static evidence

Supplied evidence

Bounded Repository Evidence

Selected source, graph, context, and retrieval provenance supplied to the workflow.

supplied context evidence

Workflow Orchestration

my-dev-kit-orchestrator

Lifecycle, readiness, and correction control

CLI-first staged workflow control for design-first coding-agent development.

Generates bounded stage prompts, tracks artifacts and readiness, preserves Semantic Continuity, routes judge findings to correction stages, derives workflow-economics evidence, and exports handoffs. It does not execute coding agents or automatically run other ecosystem tools.

Internal workflow

Start / resume run
Current stage prompt
Artifacts + readiness
Responsibility continuity
Status / check
Judge / correction routing
Export / handoff
stage prompt + acceptance contract

External implementation actor

A human or coding agent edits the target source. The ecosystem tools do not silently perform this step.

changed application

Runtime Evidence

my-frontend-observer

Browser observation and visual acceptance

Local-first browser/runtime evidence for frontend changes, comparisons, contracts, references, and correction workflows.

Captures rendered Chromium evidence, compares baselines and candidates, evaluates protected and preserved frontend contracts and reference fidelity, supports local viewer and annotation workflows, and returns bounded runtime evidence. It never edits target source.

Internal workflow

Initialize project
Capture baseline
Check candidate
Before / after comparison
Frontend contract
Reference fidelity
Viewer + correction evidence