# niravrvaghasiya/dependency-first-architect architecture

> Architecture diagram of the GitHub repository niravrvaghasiya/dependency-first-architect, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-03). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/niravrvaghasiya/dependency-first-architect
- Repository: https://github.com/niravrvaghasiya/dependency-first-architect

## Overview

Dependency-First Architect is a planning skill for coding agents: users request a plan, and the skill structures it around dependencies, early tradeoffs, a production walking skeleton, and validation gates. The repository also contains an evaluation harness that generates, blinds, judges, and aggregates plans, plus an outcomes benchmark. The README documents the skill’s capabilities; the sampled harness excerpts and source index support its main internal stage relationships. Some documented skill references and evaluation subsystems were not sampled, so those links are documented rather than source-verified.

Main entry points: [`SKILL.md`](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/SKILL.md) (Planning Skill), [`eval/dfa_eval/cli.py`](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/cli.py) (Evaluation CLI), [`eval/dfa_eval/records.py`](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/records.py) (Run Records), [`eval/dfa_eval/generate.py`](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/generate.py) (Plan Generation)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_planning["Plan Method"]
  node_skill["Planning Skill<br/>[SKILL.md]"]
  node_tradeoffs["Tradeoff Gates<br/>[SKILL.md]"]
  node_dependency_order["Dependency Order<br/>[SKILL.md]"]
  node_phase_zero["Walking Skeleton<br/>[SKILL.md]"]
  node_validation["Validation Gates<br/>[SKILL.md]"]
  node_crosscutting["Cross-cutting Risks<br/>[SKILL.md]"]
end

subgraph group_evaluation["Plan Evaluation"]
  node_eval_cli["Evaluation CLI<br/>[cli.py]"]
  node_config["Run Configuration<br/>[config.py]"]
  node_generation["Plan Generation<br/>[generate.py]"]
  node_providers["Model Providers<br/>[providers.py]"]
  node_workspace["Session Workspace<br/>[workspace.py]"]
  node_records[("Run Records<br/>[records.py]")]
  node_blinding["Plan Blinding<br/>[blinding.py]"]
  node_judging["Plan Judging<br/>[judge.py]"]
  node_aggregation["Evidence Aggregation<br/>[aggregate.py]"]
  node_schema_validation["Record Validation<br/>[schema.py]"]
  node_record_schemas["Record Schemas<br/>[schemas.py]"]
  node_stats["Statistical Analysis<br/>[stats.py]"]
end

subgraph group_benchmark["Outcome Benchmark"]
  node_outcomes_runner["Outcome Runner<br/>[runner.py]"]
  node_webhook_task["Webhook Ledger Task"]
end

node_user(("Plan Requester"))

node_user -->|"requests plan"| node_skill
node_skill -->|"frames decisions"| node_tradeoffs
node_skill -->|"orders by dependencies"| node_dependency_order
node_skill -->|"defines early slice"| node_phase_zero
node_skill -->|"sets validation gates"| node_validation
node_skill -->|"threads concerns"| node_crosscutting
node_eval_cli -->|"loads configuration"| node_config
node_eval_cli -->|"runs generation"| node_generation
node_eval_cli -->|"runs blinding"| node_blinding
node_eval_cli -->|"runs judging"| node_judging
node_eval_cli -->|"validates run files"| node_records
node_generation -->|"makes model calls"| node_providers
node_generation -->|"prepares sessions"| node_workspace
node_generation -->|"writes generation records"| node_records
node_judging -->|"makes judge calls"| node_providers
node_judging -->|"reads run records"| node_records
node_blinding -->|"reads run records"| node_records
node_aggregation -->|"reads run data"| node_records
node_aggregation -->|"computes statistics"| node_stats
node_aggregation -->|"uses run configuration"| node_config
node_schema_validation -->|"validates against"| node_record_schemas
node_outcomes_runner -.->|"runs task attempts"| node_webhook_task
node_outcomes_runner -->|"evaluates plan conditions"| node_skill

click node_skill "https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/SKILL.md"
click node_tradeoffs "https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/SKILL.md"
click node_dependency_order "https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/SKILL.md"
click node_phase_zero "https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/SKILL.md"
click node_validation "https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/SKILL.md"
click node_crosscutting "https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/SKILL.md"
click node_eval_cli "https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/cli.py"
click node_config "https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/config.py"
click node_generation "https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/generate.py"
click node_providers "https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/providers.py"
click node_workspace "https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/workspace.py"
click node_records "https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/records.py"
click node_blinding "https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/blinding.py"
click node_judging "https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/judge.py"
click node_aggregation "https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/aggregate.py"
click node_schema_validation "https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/schema.py"
click node_record_schemas "https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/schemas.py"
click node_stats "https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/stats.py"
click node_outcomes_runner "https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/outcomes/runner.py"
click node_webhook_task "https://github.com/niravrvaghasiya/dependency-first-architect/tree/main/eval/outcomes/tasks/webhook-ledger"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_skill,node_tradeoffs,node_dependency_order,node_phase_zero,node_validation,node_crosscutting toneBlue
class node_eval_cli,node_config,node_generation,node_providers,node_workspace,node_records,node_blinding,node_judging,node_aggregation,node_schema_validation,node_record_schemas,node_stats toneAmber
class node_outcomes_runner,node_webhook_task,node_user toneMint
```

## Components

### Plan Method

- [Planning Skill](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/SKILL.md): `SKILL.md`
- [Tradeoff Gates](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/SKILL.md): `SKILL.md`
- [Dependency Order](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/SKILL.md): `SKILL.md`
- [Walking Skeleton](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/SKILL.md): `SKILL.md`
- [Validation Gates](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/SKILL.md): `SKILL.md`
- [Cross-cutting Risks](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/SKILL.md): `SKILL.md`

### Plan Evaluation

- [Evaluation CLI](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/cli.py): `eval/dfa_eval/cli.py`
- [Run Configuration](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/config.py): `eval/dfa_eval/config.py`
- [Plan Generation](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/generate.py): `eval/dfa_eval/generate.py`
- [Model Providers](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/providers.py): `eval/dfa_eval/providers.py`
- [Session Workspace](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/workspace.py): `eval/dfa_eval/workspace.py`
- [Run Records](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/records.py): `eval/dfa_eval/records.py`
- [Plan Blinding](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/blinding.py): `eval/dfa_eval/blinding.py`
- [Plan Judging](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/judge.py): `eval/dfa_eval/judge.py`
- [Evidence Aggregation](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/aggregate.py): `eval/dfa_eval/aggregate.py`
- [Record Validation](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/schema.py): `eval/dfa_eval/schema.py`
- [Record Schemas](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/schemas.py): `eval/dfa_eval/schemas.py`
- [Statistical Analysis](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/dfa_eval/stats.py): `eval/dfa_eval/stats.py`

### Outcome Benchmark

- [Outcome Runner](https://github.com/niravrvaghasiya/dependency-first-architect/blob/main/eval/outcomes/runner.py): `eval/outcomes/runner.py`
- [Webhook Ledger Task](https://github.com/niravrvaghasiya/dependency-first-architect/tree/main/eval/outcomes/tasks/webhook-ledger): `eval/outcomes/tasks/webhook-ledger`

### Other components

- **Plan Requester**

## Connections

- Plan Requester → Planning Skill: requests plan (evidence: `README.md`)
- Planning Skill → Tradeoff Gates: frames decisions (evidence: `README.md`)
- Planning Skill → Dependency Order: orders by dependencies (evidence: `README.md`)
- Planning Skill → Walking Skeleton: defines early slice (evidence: `README.md`)
- Planning Skill → Validation Gates: sets validation gates (evidence: `README.md`)
- Planning Skill → Cross-cutting Risks: threads concerns (evidence: `README.md`)
- Evaluation CLI → Run Configuration: loads configuration (evidence: `eval/dfa_eval/cli.py`)
- Evaluation CLI → Plan Generation: runs generation (evidence: `eval/dfa_eval/cli.py`)
- Evaluation CLI → Plan Blinding: runs blinding (evidence: `eval/dfa_eval/cli.py`)
- Evaluation CLI → Plan Judging: runs judging (evidence: `eval/dfa_eval/cli.py`)
- Evaluation CLI → Run Records: validates run files (evidence: `eval/dfa_eval/cli.py`)
- Plan Generation → Model Providers: makes model calls (evidence: `eval/dfa_eval/generate.py`)
- Plan Generation → Session Workspace: prepares sessions (evidence: `eval/dfa_eval/generate.py`)
- Plan Generation → Run Records: writes generation records (evidence: `eval/dfa_eval/generate.py`)
- Plan Judging → Model Providers: makes judge calls (evidence: `eval/dfa_eval/judge.py`)
- Plan Judging → Run Records: reads run records (evidence: `eval/dfa_eval/judge.py`)
- Plan Blinding → Run Records: reads run records (evidence: `eval/dfa_eval/blinding.py`)
- Evidence Aggregation → Run Records: reads run data (evidence: `eval/dfa_eval/aggregate.py`)
- Evidence Aggregation → Statistical Analysis: computes statistics (evidence: `eval/dfa_eval/aggregate.py`)
- Evidence Aggregation → Run Configuration: uses run configuration (evidence: `eval/dfa_eval/aggregate.py`)
- Record Validation → Record Schemas: validates against (evidence: `eval/dfa_eval/schemas.py`)
- Outcome Runner → Webhook Ledger Task: runs task attempts
- Outcome Runner → Planning Skill: evaluates plan conditions (evidence: `README.md`)
