# bnet47/anthropicon architecture

> Architecture diagram of the GitHub repository bnet47/anthropicon, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-01). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/bnet47/anthropicon
- Repository: https://github.com/bnet47/anthropicon

## Overview

anthropicon is a Claude Code operating-process template, not an application framework. Its main workflow routes a developer from adoption or product discovery through implementation, assurance, and explicitly authorized delivery. Durable local goals and tasks support multi-step work; security hooks and canonical verification provide internal safeguards. The README documents the major skills and operating stages, while source excerpts show the local state manager, command guard, and verification hooks. Some skill and configuration internals are unsampled, so their relationships are shown only where documented.

Main entry points: [`scripts/agent.py`](https://github.com/bnet47/anthropicon/blob/main/scripts/agent.py) (Goal And Task Manager), [`.claude/skills/discover/SKILL.md`](https://github.com/bnet47/anthropicon/blob/main/.claude/skills/discover/SKILL.md) (Discover), [`.claude/skills/verify/SKILL.md`](https://github.com/bnet47/anthropicon/blob/main/.claude/skills/verify/SKILL.md) (Verify), [`.claude/skills/release/SKILL.md`](https://github.com/bnet47/anthropicon/blob/main/.claude/skills/release/SKILL.md) (Release)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_workflows["Workflows"]
  node_adoption["Adoption<br/>[adoption.md]"]
  node_discover["Discover<br/>[SKILL.md]"]
  node_initialize["Initialize<br/>[SKILL.md]"]
  node_implement["Implement<br/>[implementer.md]"]
  node_review["Review<br/>[SKILL.md]"]
  node_verify["Verify<br/>[SKILL.md]"]
  node_release["Release<br/>[SKILL.md]"]
end

subgraph group_context["Context And Planning"]
  node_orchestrate["Orchestrate<br/>[SKILL.md]"]
  node_checkpoint["Checkpoint And Resume<br/>[SKILL.md]"]
  node_agent_state["Goal And Task Manager<br/>[agent.py]"]
  node_agent_files[("Durable Run State<br/>[.agent]")]
end

subgraph group_safeguards["Safety And Assurance"]
  node_command_guard["Command Guard<br/>[guard-bash.py]"]
  node_handoff_guard["Handoff Guard<br/>[verify-subagent.py]"]
  node_stop_guard["Stop Verification Hook<br/>[verify-stop.py]"]
  node_canonical_checks["Canonical Checks<br/>[verify.py]"]
  node_authority_policy["Authority Policy<br/>[policy.json]"]
end

subgraph group_delivery["Delivery"]
end

node_developer(("Developer"))
node_claude(("Claude Code"))

node_developer -->|"adopts safely"| node_adoption
node_developer -->|"defines outcome"| node_discover
node_discover -->|"approves brief"| node_initialize
node_developer -->|"requests change"| node_implement
node_developer -->|"requests assurance"| node_review
node_developer -->|"requests checks"| node_verify
node_developer -->|"authorizes delivery"| node_release
node_claude -->|"routes work"| node_orchestrate
node_orchestrate -->|"manages goal"| node_agent_state
node_checkpoint -->|"saves progress"| node_agent_state
node_agent_state -->|"reads and writes"| node_agent_files
node_command_guard -.->|"guards tool use"| node_claude
node_handoff_guard -->|"checks active goal"| node_agent_files
node_stop_guard -->|"runs quick checks"| node_canonical_checks
node_verify -->|"runs checks"| node_canonical_checks
node_authority_policy -.->|"sets limits"| node_agent_state
node_release -->|"requires authorization"| node_developer

click node_adoption "https://github.com/bnet47/anthropicon/blob/main/docs/adoption.md"
click node_discover "https://github.com/bnet47/anthropicon/blob/main/.claude/skills/discover/SKILL.md"
click node_initialize "https://github.com/bnet47/anthropicon/blob/main/.claude/skills/init/SKILL.md"
click node_implement "https://github.com/bnet47/anthropicon/blob/main/.claude/agents/implementer.md"
click node_review "https://github.com/bnet47/anthropicon/blob/main/.claude/skills/review/SKILL.md"
click node_verify "https://github.com/bnet47/anthropicon/blob/main/.claude/skills/verify/SKILL.md"
click node_release "https://github.com/bnet47/anthropicon/blob/main/.claude/skills/release/SKILL.md"
click node_orchestrate "https://github.com/bnet47/anthropicon/blob/main/.claude/skills/orchestrate/SKILL.md"
click node_checkpoint "https://github.com/bnet47/anthropicon/blob/main/.claude/skills/checkpoint/SKILL.md"
click node_agent_state "https://github.com/bnet47/anthropicon/blob/main/scripts/agent.py"
click node_agent_files "https://github.com/bnet47/anthropicon/tree/main/.agent"
click node_command_guard "https://github.com/bnet47/anthropicon/blob/main/scripts/hooks/guard-bash.py"
click node_handoff_guard "https://github.com/bnet47/anthropicon/blob/main/scripts/hooks/verify-subagent.py"
click node_stop_guard "https://github.com/bnet47/anthropicon/blob/main/scripts/hooks/verify-stop.py"
click node_canonical_checks "https://github.com/bnet47/anthropicon/blob/main/scripts/verify.py"
click node_authority_policy "https://github.com/bnet47/anthropicon/blob/main/.agent/policy.json"

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_adoption,node_discover,node_initialize,node_implement,node_review,node_verify,node_release toneBlue
class node_orchestrate,node_checkpoint,node_agent_state,node_agent_files toneAmber
class node_command_guard,node_handoff_guard,node_stop_guard,node_canonical_checks,node_authority_policy toneMint
class node_developer,node_claude toneIndigo
```

## Components

### Workflows

- [Adoption](https://github.com/bnet47/anthropicon/blob/main/docs/adoption.md): `docs/adoption.md`
- [Discover](https://github.com/bnet47/anthropicon/blob/main/.claude/skills/discover/SKILL.md): `.claude/skills/discover/SKILL.md`
- [Initialize](https://github.com/bnet47/anthropicon/blob/main/.claude/skills/init/SKILL.md): `.claude/skills/init/SKILL.md`
- [Implement](https://github.com/bnet47/anthropicon/blob/main/.claude/agents/implementer.md): `.claude/agents/implementer.md`
- [Review](https://github.com/bnet47/anthropicon/blob/main/.claude/skills/review/SKILL.md): `.claude/skills/review/SKILL.md`
- [Verify](https://github.com/bnet47/anthropicon/blob/main/.claude/skills/verify/SKILL.md): `.claude/skills/verify/SKILL.md`
- [Release](https://github.com/bnet47/anthropicon/blob/main/.claude/skills/release/SKILL.md): `.claude/skills/release/SKILL.md`

### Context And Planning

- [Orchestrate](https://github.com/bnet47/anthropicon/blob/main/.claude/skills/orchestrate/SKILL.md): `.claude/skills/orchestrate/SKILL.md`
- [Checkpoint And Resume](https://github.com/bnet47/anthropicon/blob/main/.claude/skills/checkpoint/SKILL.md): `.claude/skills/checkpoint/SKILL.md`
- [Goal And Task Manager](https://github.com/bnet47/anthropicon/blob/main/scripts/agent.py): `scripts/agent.py`
- [Durable Run State](https://github.com/bnet47/anthropicon/tree/main/.agent): `.agent`

### Safety And Assurance

- [Command Guard](https://github.com/bnet47/anthropicon/blob/main/scripts/hooks/guard-bash.py): `scripts/hooks/guard-bash.py`
- [Handoff Guard](https://github.com/bnet47/anthropicon/blob/main/scripts/hooks/verify-subagent.py): `scripts/hooks/verify-subagent.py`
- [Stop Verification Hook](https://github.com/bnet47/anthropicon/blob/main/scripts/hooks/verify-stop.py): `scripts/hooks/verify-stop.py`
- [Canonical Checks](https://github.com/bnet47/anthropicon/blob/main/scripts/verify.py): `scripts/verify.py`
- [Authority Policy](https://github.com/bnet47/anthropicon/blob/main/.agent/policy.json): `.agent/policy.json`

### Other components

- **Developer**
- **Claude Code**

## Connections

- Developer → Adoption: adopts safely (evidence: `README.md`)
- Developer → Discover: defines outcome (evidence: `README.md`)
- Discover → Initialize: approves brief (evidence: `README.md`)
- Developer → Implement: requests change (evidence: `README.md`)
- Developer → Review: requests assurance (evidence: `README.md`)
- Developer → Verify: requests checks (evidence: `README.md`)
- Developer → Release: authorizes delivery (evidence: `README.md`)
- Claude Code → Orchestrate: routes work (evidence: `README.md`)
- Orchestrate → Goal And Task Manager: manages goal (evidence: `README.md`)
- Checkpoint And Resume → Goal And Task Manager: saves progress (evidence: `README.md`)
- Goal And Task Manager → Durable Run State: reads and writes (evidence: `scripts/agent.py`)
- Command Guard → Claude Code: guards tool use
- Handoff Guard → Durable Run State: checks active goal (evidence: `scripts/hooks/verify-subagent.py`)
- Stop Verification Hook → Canonical Checks: runs quick checks (evidence: `scripts/hooks/verify-stop.py`)
- Verify → Canonical Checks: runs checks (evidence: `README.md`)
- Authority Policy → Goal And Task Manager: sets limits
- Release → Developer: requires authorization (evidence: `README.md`)
