# aamsap/kakashi-learning architecture

> Architecture diagram of the GitHub repository aamsap/kakashi-learning, 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/aamsap/kakashi-learning
- Repository: https://github.com/aamsap/kakashi-learning

## Overview

Kakashi Learning is a Hermes skill for researching repository methods and tailoring a skill, a patch, or a learning note to a user's constraints. Its principal workflow establishes context and success criteria, performs bounded discovery, evaluates pinned evidence, makes reuse decisions, drafts an artifact, and evaluates it before handoff. An optional Python helper checks ledger structure and recorded readiness offline; it does not perform research, run tests, or install artifacts. The workflow and helper are documented, but the skill source itself was not excerpted, so internal calls are not inferred.

Main entry points: [`kakashi-learning/SKILL.md`](https://github.com/aamsap/kakashi-learning/blob/main/kakashi-learning/SKILL.md) (Behavioral evaluation), [`scripts/audit.py`](https://github.com/aamsap/kakashi-learning/blob/main/scripts/audit.py) (Ledger auditor), [`templates/ledger.json`](https://github.com/aamsap/kakashi-learning/blob/main/templates/ledger.json) (Machine ledger), [`evals/README.md`](https://github.com/aamsap/kakashi-learning/blob/main/evals/README.md) (Evaluation protocol)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_learning["Learning workflow"]
  node_skill["Kakashi skill<br/>[SKILL.md]"]
  node_context["Context and goals<br/>[SKILL.md]"]
  node_discovery["Bounded discovery<br/>[SKILL.md]"]
  node_evidence["Pinned evidence<br/>[SKILL.md]"]
  node_decisions["Reuse decisions<br/>[SKILL.md]"]
  node_draft["Tailored artifact<br/>[SKILL.md]"]
  node_evaluation["Behavioral evaluation<br/>[SKILL.md]"]
  node_handoff["Approval handoff<br/>[SKILL.md]"]
end

subgraph group_records["Research records"]
  node_report["Human report<br/>[report.md]"]
  node_ledger["Machine ledger<br/>[ledger.json]"]
  node_eval_protocol["Evaluation protocol<br/>[README.md]"]
end

subgraph group_checks["Offline checks"]
  node_audit["Ledger auditor<br/>[audit.py]"]
end

node_user(("Hermes user"))

node_user -->|"sets goal"| node_skill
node_skill -->|"establishes"| node_context
node_context -->|"guides"| node_discovery
node_discovery -->|"pins sources"| node_evidence
node_evidence -->|"informs"| node_decisions
node_decisions -->|"tailors"| node_draft
node_draft -->|"evaluates"| node_evaluation
node_evaluation -->|"reports results"| node_handoff
node_user -->|"records run"| node_report
node_user -->|"creates run"| node_ledger
node_eval_protocol -->|"defines protocol"| node_evaluation
node_user -->|"runs checks"| node_audit
node_audit -->|"validates"| node_ledger

click node_skill "https://github.com/aamsap/kakashi-learning/blob/main/kakashi-learning/SKILL.md"
click node_context "https://github.com/aamsap/kakashi-learning/blob/main/kakashi-learning/SKILL.md"
click node_discovery "https://github.com/aamsap/kakashi-learning/blob/main/kakashi-learning/SKILL.md"
click node_evidence "https://github.com/aamsap/kakashi-learning/blob/main/kakashi-learning/SKILL.md"
click node_decisions "https://github.com/aamsap/kakashi-learning/blob/main/kakashi-learning/SKILL.md"
click node_draft "https://github.com/aamsap/kakashi-learning/blob/main/kakashi-learning/SKILL.md"
click node_evaluation "https://github.com/aamsap/kakashi-learning/blob/main/kakashi-learning/SKILL.md"
click node_handoff "https://github.com/aamsap/kakashi-learning/blob/main/kakashi-learning/SKILL.md"
click node_report "https://github.com/aamsap/kakashi-learning/blob/main/templates/report.md"
click node_ledger "https://github.com/aamsap/kakashi-learning/blob/main/templates/ledger.json"
click node_eval_protocol "https://github.com/aamsap/kakashi-learning/blob/main/evals/README.md"
click node_audit "https://github.com/aamsap/kakashi-learning/blob/main/scripts/audit.py"

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_context,node_discovery,node_evidence,node_decisions,node_draft,node_evaluation,node_handoff,node_user toneBlue
class node_report,node_ledger,node_eval_protocol toneAmber
class node_audit toneMint
```

## Components

### Learning workflow

- [Kakashi skill](https://github.com/aamsap/kakashi-learning/blob/main/kakashi-learning/SKILL.md): `kakashi-learning/SKILL.md`
- [Context and goals](https://github.com/aamsap/kakashi-learning/blob/main/kakashi-learning/SKILL.md): `kakashi-learning/SKILL.md`
- [Bounded discovery](https://github.com/aamsap/kakashi-learning/blob/main/kakashi-learning/SKILL.md): `kakashi-learning/SKILL.md`
- [Pinned evidence](https://github.com/aamsap/kakashi-learning/blob/main/kakashi-learning/SKILL.md): `kakashi-learning/SKILL.md`
- [Reuse decisions](https://github.com/aamsap/kakashi-learning/blob/main/kakashi-learning/SKILL.md): `kakashi-learning/SKILL.md`
- [Tailored artifact](https://github.com/aamsap/kakashi-learning/blob/main/kakashi-learning/SKILL.md): `kakashi-learning/SKILL.md`
- [Behavioral evaluation](https://github.com/aamsap/kakashi-learning/blob/main/kakashi-learning/SKILL.md): `kakashi-learning/SKILL.md`
- [Approval handoff](https://github.com/aamsap/kakashi-learning/blob/main/kakashi-learning/SKILL.md): `kakashi-learning/SKILL.md`

### Research records

- [Human report](https://github.com/aamsap/kakashi-learning/blob/main/templates/report.md): `templates/report.md`
- [Machine ledger](https://github.com/aamsap/kakashi-learning/blob/main/templates/ledger.json): `templates/ledger.json`
- [Evaluation protocol](https://github.com/aamsap/kakashi-learning/blob/main/evals/README.md): `evals/README.md`

### Offline checks

- [Ledger auditor](https://github.com/aamsap/kakashi-learning/blob/main/scripts/audit.py): `scripts/audit.py`

### Other components

- **Hermes user**

## Connections

- Hermes user → Kakashi skill: sets goal (evidence: `README.md`)
- Kakashi skill → Context and goals: establishes (evidence: `README.md`)
- Context and goals → Bounded discovery: guides (evidence: `README.md`)
- Bounded discovery → Pinned evidence: pins sources (evidence: `README.md`)
- Pinned evidence → Reuse decisions: informs (evidence: `README.md`)
- Reuse decisions → Tailored artifact: tailors (evidence: `README.md`)
- Tailored artifact → Behavioral evaluation: evaluates (evidence: `README.md`)
- Behavioral evaluation → Approval handoff: reports results (evidence: `README.md`)
- Hermes user → Human report: records run (evidence: `README.md`)
- Hermes user → Machine ledger: creates run (evidence: `README.md`)
- Evaluation protocol → Behavioral evaluation: defines protocol (evidence: `README.md`)
- Hermes user → Ledger auditor: runs checks (evidence: `README.md`)
- Ledger auditor → Machine ledger: validates (evidence: `README.md`)
