# tjboudreaux/cc-thinking-skills architecture

> Architecture diagram of the GitHub repository tjboudreaux/cc-thinking-skills, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-05). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/tjboudreaux/cc-thinking-skills
- Repository: https://github.com/tjboudreaux/cc-thinking-skills

## Overview

This repository is a portable catalog of structured-reasoning Agent Skills. A user invokes the router when the right framework is unclear; it selects no skill, one skill, or a small set, while users can also invoke a matching leaf skill directly. The `evals/` tree is a substantial supporting product branch: dataset preparation feeds evaluation runners, which use model execution and scoring utilities to produce evidence. The graph reflects documented skill behavior and indexed runner dependencies; unsampled evaluation modules are represented only where their paths and roles are supported.

Main entry points: [`evals/lib/objective.js`](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/objective.js) (Study execution), [`evals/run-routing-boundary.js`](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/run-routing-boundary.js) (Routing boundary study), [`evals/lib/droid.js`](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/droid.js) (Droid model execution), [`evals/run-scientific-method-experiment.js`](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/run-scientific-method-experiment.js) (Scientific-method study)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_catalog["Skill catalog"]
  node_router["Model router<br/>[SKILL.md]"]
  node_leafskills["Reasoning skills"]
end

subgraph group_evaluation["Evaluation studies"]
  node_datasetprep["Dataset preparation"]
  node_swebench["SWE-bench mapping<br/>[swebench-pro.js]"]
  node_routingrun["Routing evaluation<br/>[run-routing.js]"]
  node_boundaryrun["Routing boundary study"]
  node_calibrationrun["Calibration runner<br/>[run-calibration.js]"]
  node_objectiverun["Objective runner<br/>[run-objective.js]"]
  node_pairwiserun["Pairwise runner<br/>[run-pairwise.js]"]
  node_scirun["Scientific-method study"]
  node_otherstudies["Other eval studies"]
end

subgraph group_runtime["Evaluation runtime"]
  node_skillsloader["Skill catalog loader<br/>[skills.js]"]
  node_modelexec["Droid model execution<br/>[droid.js]"]
  node_studyengine["Study execution<br/>[objective.js]"]
  node_judge["Pairwise judging<br/>[judge.js]"]
  node_resultmodel["Result envelopes<br/>[result.js]"]
  node_io["Run I/O<br/>[io.js]"]
  node_stats["Statistics<br/>[stats.js]"]
  node_conditions["Study conditions<br/>[conditions.js]"]
end

subgraph group_evidence["Evidence and analysis"]
  node_evidence["Evidence registry<br/>[evidence.js]"]
  node_registry[("Evidence registry data<br/>[registry.js]")]
  node_portfolio["Portfolio analysis<br/>[portfolio.js]"]
  node_analysis["Published audit"]
end

node_user(("Agent user"))

node_user -->|"invokes when unsure"| node_router
node_user -->|"invokes matching skill"| node_leafskills
node_router -->|"selects framework"| node_leafskills
node_datasetprep -->|"maps issue rows"| node_swebench
node_routingrun -->|"loads skill catalog"| node_skillsloader
node_routingrun -->|"queries model"| node_modelexec
node_routingrun -->|"writes run output"| node_io
node_boundaryrun -->|"loads skill catalog"| node_skillsloader
node_boundaryrun -->|"queries model"| node_modelexec
node_boundaryrun -->|"orders study arms"| node_studyengine
node_boundaryrun -->|"hashes artifacts"| node_resultmodel
node_boundaryrun -->|"writes artifacts"| node_io
node_calibrationrun -->|"runs model calls"| node_modelexec
node_calibrationrun -->|"uses run I/O"| node_io
node_objectiverun -.->|"runs objective study"| node_studyengine
node_pairwiserun -->|"runs model calls"| node_modelexec
node_pairwiserun -->|"uses judge panel"| node_judge
node_pairwiserun -->|"uses result model"| node_resultmodel
node_scirun -->|"runs model calls"| node_modelexec
node_scirun -->|"analyzes objective"| node_studyengine
node_scirun -->|"uses result model"| node_resultmodel
node_scirun -->|"reads and writes"| node_io
node_otherstudies -.->|"prepares evaluation data"| node_datasetprep
node_studyengine -->|"creates result envelopes"| node_resultmodel
node_studyengine -->|"uses study conditions"| node_conditions
node_studyengine -->|"uses statistics"| node_stats
node_judge -->|"requests judge votes"| node_modelexec
node_evidence -->|"uses registry"| node_registry
node_analysis -.->|"reports evidence"| node_evidence
node_analysis -.->|"summarizes portfolio"| node_portfolio

click node_router "https://github.com/tjboudreaux/cc-thinking-skills/blob/main/skills/thinking-model-router/SKILL.md"
click node_leafskills "https://github.com/tjboudreaux/cc-thinking-skills/tree/main/skills"
click node_datasetprep "https://github.com/tjboudreaux/cc-thinking-skills/tree/main/evals/datasets"
click node_swebench "https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/datasets/swebench-pro.js"
click node_routingrun "https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/run-routing.js"
click node_boundaryrun "https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/run-routing-boundary.js"
click node_calibrationrun "https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/run-calibration.js"
click node_objectiverun "https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/run-objective.js"
click node_pairwiserun "https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/run-pairwise.js"
click node_scirun "https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/run-scientific-method-experiment.js"
click node_otherstudies "https://github.com/tjboudreaux/cc-thinking-skills/tree/main/evals"
click node_skillsloader "https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/skills.js"
click node_modelexec "https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/droid.js"
click node_studyengine "https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/objective.js"
click node_judge "https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/judge.js"
click node_resultmodel "https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/result.js"
click node_io "https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/io.js"
click node_stats "https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/stats.js"
click node_conditions "https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/conditions.js"
click node_evidence "https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/evidence.js"
click node_registry "https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/registry.js"
click node_portfolio "https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/portfolio.js"
click node_analysis "https://github.com/tjboudreaux/cc-thinking-skills/tree/main/analysis"

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_router,node_leafskills,node_user toneBlue
class node_datasetprep,node_swebench,node_routingrun,node_boundaryrun,node_calibrationrun,node_objectiverun,node_pairwiserun,node_scirun,node_otherstudies toneAmber
class node_skillsloader,node_modelexec,node_studyengine,node_judge,node_resultmodel,node_io,node_stats,node_conditions toneMint
class node_evidence,node_registry,node_portfolio,node_analysis toneRose
```

## Components

### Skill catalog

- [Model router](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/skills/thinking-model-router/SKILL.md): `skills/thinking-model-router/SKILL.md`
- [Reasoning skills](https://github.com/tjboudreaux/cc-thinking-skills/tree/main/skills): `skills`

### Evaluation studies

- [Dataset preparation](https://github.com/tjboudreaux/cc-thinking-skills/tree/main/evals/datasets): `evals/datasets`
- [SWE-bench mapping](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/datasets/swebench-pro.js): `evals/datasets/swebench-pro.js`
- [Routing evaluation](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/run-routing.js): `evals/run-routing.js`
- [Routing boundary study](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/run-routing-boundary.js): `evals/run-routing-boundary.js`
- [Calibration runner](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/run-calibration.js): `evals/run-calibration.js`
- [Objective runner](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/run-objective.js): `evals/run-objective.js`
- [Pairwise runner](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/run-pairwise.js): `evals/run-pairwise.js`
- [Scientific-method study](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/run-scientific-method-experiment.js): `evals/run-scientific-method-experiment.js`
- [Other eval studies](https://github.com/tjboudreaux/cc-thinking-skills/tree/main/evals): `evals`

### Evaluation runtime

- [Skill catalog loader](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/skills.js): `evals/lib/skills.js`
- [Droid model execution](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/droid.js): `evals/lib/droid.js`
- [Study execution](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/objective.js): `evals/lib/objective.js`
- [Pairwise judging](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/judge.js): `evals/lib/judge.js`
- [Result envelopes](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/result.js): `evals/lib/result.js`
- [Run I/O](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/io.js): `evals/lib/io.js`
- [Statistics](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/stats.js): `evals/lib/stats.js`
- [Study conditions](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/conditions.js): `evals/lib/conditions.js`

### Evidence and analysis

- [Evidence registry](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/evidence.js): `evals/lib/evidence.js`
- [Evidence registry data](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/registry.js): `evals/lib/registry.js`
- [Portfolio analysis](https://github.com/tjboudreaux/cc-thinking-skills/blob/main/evals/lib/portfolio.js): `evals/lib/portfolio.js`
- [Published audit](https://github.com/tjboudreaux/cc-thinking-skills/tree/main/analysis): `analysis`

### Other components

- **Agent user**

## Connections

- Agent user → Model router: invokes when unsure (evidence: `README.md`)
- Agent user → Reasoning skills: invokes matching skill (evidence: `README.md`)
- Model router → Reasoning skills: selects framework (evidence: `README.md`)
- Dataset preparation → SWE-bench mapping: maps issue rows (evidence: `evals/datasets/ingest-hf.js`)
- Routing evaluation → Skill catalog loader: loads skill catalog (evidence: `evals/run-routing.js`)
- Routing evaluation → Droid model execution: queries model (evidence: `evals/run-routing.js`)
- Routing evaluation → Run I/O: writes run output (evidence: `evals/run-routing.js`)
- Routing boundary study → Skill catalog loader: loads skill catalog (evidence: `evals/run-routing-boundary.js`)
- Routing boundary study → Droid model execution: queries model (evidence: `evals/run-routing-boundary.js`)
- Routing boundary study → Study execution: orders study arms (evidence: `evals/run-routing-boundary.js`)
- Routing boundary study → Result envelopes: hashes artifacts (evidence: `evals/run-routing-boundary.js`)
- Routing boundary study → Run I/O: writes artifacts (evidence: `evals/run-routing-boundary.js`)
- Calibration runner → Droid model execution: runs model calls (evidence: `evals/run-calibration.js`)
- Calibration runner → Run I/O: uses run I/O (evidence: `evals/run-calibration.js`)
- Objective runner → Study execution: runs objective study
- Pairwise runner → Droid model execution: runs model calls (evidence: `evals/run-pairwise.js`)
- Pairwise runner → Pairwise judging: uses judge panel (evidence: `evals/run-pairwise.js`)
- Pairwise runner → Result envelopes: uses result model (evidence: `evals/run-pairwise.js`)
- Scientific-method study → Droid model execution: runs model calls (evidence: `evals/run-scientific-method-experiment.js`)
- Scientific-method study → Study execution: analyzes objective (evidence: `evals/run-scientific-method-experiment.js`)
- Scientific-method study → Result envelopes: uses result model (evidence: `evals/run-scientific-method-experiment.js`)
- Scientific-method study → Run I/O: reads and writes (evidence: `evals/run-scientific-method-experiment.js`)
- Other eval studies → Dataset preparation: prepares evaluation data (evidence: `evals/datasets/ingest-hf.js`)
- Study execution → Result envelopes: creates result envelopes (evidence: `evals/lib/objective.js`)
- Study execution → Study conditions: uses study conditions (evidence: `evals/lib/objective.js`)
- Study execution → Statistics: uses statistics (evidence: `evals/lib/objective.js`)
- Pairwise judging → Droid model execution: requests judge votes (evidence: `evals/lib/judge.js`)
- Evidence registry → Evidence registry data: uses registry (evidence: `evals/lib/evidence.js`)
- Published audit → Evidence registry: reports evidence
- Published audit → Portfolio analysis: summarizes portfolio
