JOSHUA BLOODWORTH
Agent-readable portfolio · static HTML

Joshua Bloodworth

My work focuses on AI enablement and practical AI workflows.

Browse the complete project catalog in readable HTML. Some projects have public resources; others have a description of the work.

Read the resume text · Open the complete resume PDF · Plain text index for LLMs

Products & systems

Products & systems

Tools built around real work, with the decisions left visible.

AI system · Mac app available as a prerelease.

DUKE Autorouter

Choose an AI model for a task, then check its answer.

Topics: Agents & permissions · Evidence & evaluation

Available: Source code · Screenshots

Problem
Choosing a model and deciding how much effort a task needs can become a task of its own.
Built
A Mac app that routes tasks across a chosen set of AI models, reviews their answers, and makes a limited number of corrections when needed.
Design decision
I kept the routing choice and review checks visible, with a limit on automated correction attempts.
Where it stands
This sample produced the requested guide; two automated checks remained uncertain. A generated answer still needs human judgment.
Take a look
Browse the source code, download the Mac app, and explore its routing checks on GitHub.

Explore the app and source code ↗

View visual references ↗

Product build · Desktop and server versions available in alpha.

Band Office

Keep the moving parts of a school music program in one place.

Topics: Workflows & operations

Available: Source code · Screenshots

Problem
Directors juggle students, instruments, checkouts, events, and records across disconnected systems.
Built
A desktop app and self-hosted option that bring a music program’s day-to-day work together.
Design decision
I brought students, instruments, checkouts, and events into one working record system, with desktop and self-hosted options.
Where it stands
The alpha releases and sample screens are available to inspect. Reliability in a school’s day-to-day use still needs evaluation.
Take a look
Explore the source code, sample-data screenshots, and desktop and server releases.

Explore Band Office ↗

View the visual reference ↗

Personal operating system · Local operating system · related sample files available

The Desk / AgentOS

A file-backed system for planning work, keeping decisions, and making projects easier to resume.

Topics: Agents & permissions · Workflows & operations

Available: Sample files

Problem
Long-running work needs a place for priorities, approvals, and current project context.
Built
A local operating layer with structured project records, work queues, handoffs, and a supporting cockpit.
Where it stands
These public sample files illustrate the operating-file structure. They are not the private Desk cockpit or its live work queues.
Take a look
Inspect the fictional operating files included with the related AgentOS Doctor tool.

Inspect the related operating-file example ↗

Review workspace · Local review workspace · screenshot and sample input available

RoleProof

Review one evidence-backed record at a time, with a clear history of human decisions.

Topics: Evidence & evaluation · Workflows & operations

Available: Screenshots · Sample files

Problem
When source material changes, a prior decision may need to be reconsidered.
Built
A local review workspace that preserves intake snapshots, records reviews, and produces exports after human approval.
Design decision
I preserved source snapshots and human review decisions separately so a changed record can be reconsidered.
Where it stands
This example shows intake and review controls, rather than a completed approval or export. The practice build is independent of the live job-search system; all names, URLs, and salaries in the sample are fictional.
Take a look
Inspect the review interface and fifty fictional job records from the practice build’s deterministic generator, including missing fields and deliberately invalid examples.

Inspect the fictional intake sample ↗

View the visual reference ↗

Routing utility · Public source package

RouteWitness

A small utility for choosing a model route and keeping a record of the choice.

Topics: Agents & permissions

Available: Source code · Sample files

Problem
Routing decisions need consistent rules and a way to explain why a route was chosen.
Built
A provider-native routing policy with capability and effort recommendations, conservative risk flags, and metadata-only receipts that distinguish a planned route from witnessed execution.
Where it stands
The package is advisory and experimental. It does not switch an active root conversation, proxy traffic across providers, or establish model quality or savings.
Take a look
Browse the public source package, documentation, and available examples.

Explore the public repository ↗

Inspect the public examples ↗

Data-model toolkit · Local toolkit · sample JSON available

TableForge

Define and check the structure of an application’s data before building its interface.

Topics: Workflows & operations

Available: Sample files

Problem
Unclear tables, linked records, and field rules can make an application difficult to change.
Built
A toolkit for validating tables, records, linked fields, and views, with sample schemas for practical workflows.
Design decision
I made field and record rules inspectable before building screens on top of them.
Where it stands
These are data-contract examples. The record’s example.com link and sample score are placeholders.
Take a look
Open the original Proof Library schema and one fictional record. The schema defines required fields, choices, a filtered view, and sorting.

Inspect the sample schema ↗

Inspect the sample records ↗

Application planning toolkit · Local planning toolkit · sample JSON available

AppForge

Turn a checked data model into an inspectable application plan.

Topics: Workflows & operations

Available: Sample files

Problem
A data schema does not explain which screens people need or how they will use them.
Built
Structured application templates and screen descriptions that extend the TableForge data contract.
Design decision
I carried field bindings, view filters, and sorting from the data contract into an explicit screen plan.
Where it stands
The output is an application plan in JSON. It does not generate a completed frontend or deployed application.
Take a look
Inspect the original template and screen manifest produced for the same Proof Library data contract.

Inspect the application template ↗

Inspect the screen manifest ↗

Business workflow system · Local build · sample schema available

Gig City CRM

Organize business contacts, companies, and sales work in a shared application.

Topics: Workflows & operations

Available: Sample files

Problem
Client and prospect information becomes difficult to use when spread across separate notes and tools.
Built
A client-management system with structured contacts and company records, plus focused workflow tools.
Where it stands
This reference is a data contract for the CRM workflow, not a screenshot of customer records or proof of a production deployment.
Take a look
Inspect the original TableForge data contract for companies, contacts, opportunities, activities, and tasks.

Inspect the CRM data contract ↗

Substitute lesson-planning tool · Live web tool

Gig City Sub-Save

Prepare an emergency substitute lesson plan for a chosen subject, grade, and class period.

Topics: Learning & enablement

Available: Live site · Screenshots

Problem
A substitute needs usable teaching instructions when a teacher has little time to prepare.
Built
A web tool with subject, grade, duration, substitute-experience, topic, and optional Tennessee-standards inputs, plus options for a read-aloud script and student activity sheet.
Where it stands
The live interface is available. This review did not evaluate generated lessons or establish school-wide use or learning outcomes.
Take a look
Open the live planning form and inspect the interface screenshot.

Open Sub-Save ↗

View the visual reference ↗

Local publishing utility · Local generator · generated example available

Morning Dispatch

Turn a small source-data file into a printable morning newspaper.

Topics: Workflows & operations

Available: Sample files

Problem
A useful personal brief needs a readable structure before adding integrations or automation.
Built
A local generator that normalizes manually supplied JSON into newspaper sections and printable HTML.
Where it stands
This example uses manual fictional JSON. Live email, calendar, and automatic printing integrations are not demonstrated.
Take a look
Open a newspaper generated by the original CLI from the supplied fictional workshop input.

Read the generated sample newspaper ↗

Inspect its fictional source input ↗

Interactive portfolio · Published portfolio · 3D and readable views

The Museum of Possibility

Explore AI implementation and enablement work through a connected 3D museum.

Topics: Websites & portfolios

Available: Live site · Screenshots

Problem
A portfolio needs to show both the work and the person behind it, while giving visitors a quick way to find relevant examples.
Built
A browser-based museum with six themed exhibits, an illustrated visitor, a retro desktop catalog, a playable resume, and readable project pages.
Design decision
I made exploration optional: a project plaque opens the actual work, and visitors can use the Archives, readable pages, or resume directly.
Where it stands
The museum uses original exhibit concepts, generated artwork and a textured visitor model, rendered in Three.js. The 3D experience is heavier than the text view.
Take a look
Explore the museum, browse its project catalog, or use the text view.

Explore the museum ↗

Browse the readable catalog ↗

View the visual reference ↗

Operating-file diagnostic · Public source package

AgentOS Doctor

Check whether an AI operating system’s files, state, dashboards, and governance have drifted apart.

Topics: Agents & permissions · Evidence & evaluation

Available: Source code · Sample files

Problem
A long-running operating system can keep showing inconsistent priorities or stale state across its files.
Built
A report-only skill for Codex and Claude Code, with configurable operating-file checks and a fictional sample workspace.
Where it stands
The diagnostic reports mismatches for human review. It does not edit files, choose priorities, or change authority.
Take a look
Browse the public source, documentation, and included examples.

Explore the repository ↗

Inspect the fictional operating-file examples ↗

Software handoff toolkit · Public source package

Small Software Capsule

Keep a small tool’s purpose, launch command, access boundaries, and health check beside it.

Topics: Releases & handoffs

Available: Source code · Sample files

Problem
Months after a small tool was built, its purpose and safe operating conditions can be difficult to recover.
Built
A plain capsule.json manifest and Python CLI for manifest validation, path checks, and readable health reports.
Where it stands
Health commands run only when explicitly requested. A passing report is not a security audit or a guarantee that the software is safe.
Take a look
Browse the public source, documentation, and included examples.

Explore the repository ↗

Inspect the example capsule ↗

Permission-aware AI workbench · Public source package

AcmeOS: Governed AI Workbench

Answer a customer question using permitted company sources, then require approval before acting.

Topics: Agents & permissions · Workflows & operations

Available: Source code · Screenshots · Visual preview

Problem
A useful answer needs to respect role access, and a draft must remain separate from permission to send it.
Built
A React workbench and Python backend with role-filtered sources, citations, owner approvals, local dry-run outbox artifacts, and event receipts.
Where it stands
The company and customer are fictional. Role selection is a demonstration control, not authentication; production identity, multitenancy, and live email sending are out of scope.
Take a look
Browse the public source, documentation, and included examples.

Explore the repository ↗

Watch the recorded workflow ↗

View the visual reference ↗

Content workflow system · Local workflow build · screenshot available

Gig City Content OS

Move content from an initial idea through drafting, review, and planning.

Topics: Workflows & operations

Available: Screenshots

Problem
A draft and an approved version need to remain distinguishable as content changes.
Built
A content workspace with revision-specific approval and a persistent record of work.
Design decision
I tied approval to a particular revision so an edited draft would not silently inherit a previous approval.
Where it stands
This reference shows the local demo interface. It does not demonstrate an automated publishing run or publishing results.
Take a look
Inspect the demo dashboard and its separate queues for shaping, approval, review, and planning.

View the visual reference ↗

Workflow demonstration · Local demonstration · screenshot available

AI Implementation Desk

Explore AI-supported implementation work through a fictional field-operations company.

Topics: Learning & enablement · Workflows & operations

Available: Screenshots

Problem
Implementation teams need answers, readiness decisions, and review points tied to supplied source material.
Built
A synthetic-data demonstration with source-based answers, readiness briefs, recommendation review, and an audit history.
Design decision
I separated source-backed answers, readiness recommendations, human approval, and the audit history into distinct stages.
Where it stands
This build uses synthetic documents, deterministic retrieval and generation, and local browser state. The screenshot demonstrates the workflow, not customer deployment.
Take a look
Inspect the dashboard screenshot from the fictional Northstar FieldOps demonstration.

View the visual reference ↗

Hiring workflow demonstration · Local build · screenshot available

Slatekeeper

Prepare hiring materials while keeping candidate decisions with a human reviewer.

Topics: Workflows & operations

Available: Screenshots

Problem
Hiring support needs to distinguish supplied candidate evidence from generated drafts.
Built
A fictional-role demonstration with a hiring brief, duty scorecard, candidate packet, interview guide, debrief, and communication drafts. Its agent behavior is simulated.
Where it stands
Agent behavior is simulated. The sample is a hiring workflow demonstration, not an automated candidate-selection system.
Take a look
Inspect the original interface capture.

View the visual reference ↗

Planning-tool suite · Local build · screenshot available

AI Enablement Workflow Suite

Explore workflow opportunities, adoption planning, and workforce readiness together.

Topics: Learning & enablement · Workflows & operations

Available: Screenshots

Problem
A promising AI task also needs stakeholder planning, review, and preparation for the people using it.
Built
Three planning demonstrations: Automation ROI Console, AI Adoption Playbook, and AI Training Diagnostic. Scoring and planning are deterministic; projected savings are not demonstrated results.
Where it stands
These three frontend tools use deterministic planning and scoring. They do not establish adoption, customer use, or measured savings.
Take a look
Inspect the original interface captures.

View visual references ↗

Workshops & guides

Workshops & guides

Ways to help people understand a process and use AI with judgment.

Workshop · Workshop materials available.

Process Archaeology

Uncover how work really gets done before automating it.

Topics: Workflows & operations · Learning & enablement

Available: Download · Visual preview

Problem
The written process rarely captures every handoff, exception, or decision people make along the way.
Built
A workshop with guided exercises, facilitator slides, and a companion field kit for turning unwritten know-how into a guide a team can test.
Design decision
I put discovery before automation: participants capture the exceptions and human decisions, then turn them into a process someone else can test.
Where it stands
The deck and field kit are complete materials. Their usefulness in a particular team depends on the process and the people participating.
Take a look
Read the 30-slide workshop deck and 15-page companion field kit.

Open participant field kit ↗

Open workshop slides ↗

View the visual reference ↗

Framework document · Framework handout

PROVE

A two-page framework for introducing AI into a workflow with clear decisions and review points.

Topics: Learning & enablement · Workflows & operations

Available: Download · Visual preview

Problem
Teams need a practical way to decide where AI fits, who is responsible, and what to check before expanding its use.
Built
A framework handout covering the stages, decisions, and checks involved in AI implementation.
Take a look
Read the two-page framework PDF.

Open framework PDF ↗

View the visual reference ↗

Practical AI guide · Published guide · free to read.

Taking AI to Work

A practical field guide to using AI in everyday work.

Topics: Learning & enablement

Available: Live site · Visual preview

Problem
People need concrete examples of what to give AI, what to ask it to do, and how to check the result.
Built
A 382-page guide with worked examples across business functions, plus companion materials to use alongside them.
Design decision
I used a fictional company to make workplace examples concrete while keeping inputs, instructions, and review steps together.
Where it stands
The complete guide is published. Its examples illustrate approaches to adapt and check in your own context.
Take a look
Read the complete guide and explore its companion resources online.

Read the guide ↗

View the visual reference ↗

Book and workbook · Published book and workbook · free to read.

Making Room for AI

Help a team turn one useful AI task into a new way of working.

Topics: Learning & enablement · Workflows & operations

Available: Live site · Visual preview

Problem
Adding AI to a task raises practical questions: who has access, who reviews the result, and what does the team do with the time it creates?
Built
A 118-page book for team leaders, with a 31-page workbook on introducing AI, helping people use it, and reconsidering how work gets done.
Design decision
I followed the change around an AI task: access, responsibilities, human review, and what a team does with the capacity it creates.
Where it stands
The book and workbook are published. The team story is fictional; it does not establish measured savings or an adoption outcome.
Take a look
Read the complete book and download the workbook.

Read the book and workbook ↗

View the visual reference ↗

AI learning platform · Published courses · free previews; full Builds are paid.

The Practical AI Workbench

Learn practical AI skills by building something useful.

Topics: Learning & enablement

Available: Live site · Visual preview

Problem
Knowing what an AI tool can do is different from knowing how to use it well in your own work.
Built
Six self-paced courses, called Builds, covering reusable prompts, context, personal work systems, model selection, agents, and repeatable workflows.
Design decision
I organized learning around a useful thing the learner makes, with guided activities and materials they can return to.
Where it stands
Six preview lessons are open to browse. Full Builds are paid; completing a lesson is a starting point for applying and checking the skill.
Take a look
Try six free preview lessons. Full Builds include guided activities and downloadable materials.

Explore the free preview lessons ↗

View the visual reference ↗

Workflow worksheet · Free worksheet · PDF and browser versions

Find the Five Hours

Find recurring work that may be worth simplifying or redesigning.

Topics: Workflows & operations

Available: Live site · Download · Screenshots

Problem
It can be difficult to choose a first AI workflow without understanding where time is going.
Built
A Practical AI Workbench worksheet for examining recurring tasks and choosing a bounded starting point.
Take a look
Use the four-page worksheet in your browser or open the printable PDF. Five hours is a search target, not a guaranteed saving.

Open the worksheet PDF ↗

Use the fillable browser worksheet ↗

View the visual reference ↗

Readiness guide · Public source package

Knowledge Readiness Field Kit

Examine the information and decisions a workflow needs before adding AI.

Topics: Evidence & evaluation · Learning & enablement

Available: Source code · Sample files

Problem
An AI-enabled change depends on usable information, clear decisions, and a realistic starting point.
Built
A command-line field kit that classifies prepared knowledge and writes a readiness manifest, handoff-safety notes, and proof cases.
Where it stands
Classification depends on supplied source records and policy. A readiness packet does not independently validate a disputed fact.
Take a look
Browse the public source package, documentation, and available examples.

Explore the public repository ↗

Inspect the public examples ↗

Practical AI field guide · Published guide · free downloads

Ok, Boomer

Learn to use AI for everyday tasks while keeping your own judgment.

Topics: Learning & enablement

Available: Live site · Download

Problem
People need approachable examples and a clear way to check an AI answer, without a technical background.
Built
A 164-page illustrated field guide with 14 practical playbooks and an Ask, Notice, Decide routine. Boomer and Scout guide readers through ordinary tasks, checking answers, and deciding what to use.
Take a look
Read the complete guide for free in PDF, EPUB, or Word. No account or signup required.

Visit the field guide ↗

Download PDF ↗

Download EPUB ↗

Download Word ↗

Interactive learning

Interactive learning

Practice experiences that make complex ideas easier to work through.

Practice game · Local study game · screenshot and questions available

Architecture Rescue

A study game with 60 original questions and ten-question repair rounds.

Topics: Learning & enablement

Available: Screenshots · Sample files

Problem
Technical exam preparation benefits from practice that explains decisions rather than only marking answers.
Built
A browser game with explanations, retries, progress export, and an animated machine assembled through correct answers.
Design decision
I made each answer explain the architectural choice and connect it to the machine-repair progression.
Where it stands
These are independently written practice questions. The game is not an official exam or a validated measure of exam readiness.
Take a look
Inspect the repair game screenshot and an earlier seventy-five-question study bank with choices, explanations, objective mapping, and source references.

Inspect the earlier study question bank ↗

View the visual reference ↗

Practice workflow build · Local API exercise · sample input available

Readiness Lab: Retail Returns

Check the information available in fictional retail-return records.

Topics: Workflows & operations

Available: Sample files

Problem
Before using workflow data, it helps to know which information is present and which is missing.
Built
A local Python API practice build that checks basic data coverage for fictional return records and keeps the source row number in each result.
Where it stands
The API reports input coverage and preserves source rows. It does not make refund decisions or establish operational readiness.
Take a look
Inspect the two fictional input records used by the API’s source-row and missing-reason test.

Inspect the fictional return records ↗

Interactive simulation · Local simulations · screenshot available

Causal Contraption

Browser simulations for testing how changes in one part of a system affect another.

Topics: Creative explorations

Available: Screenshots

Problem
Causal relationships are hard to grasp when presented as static diagrams.
Built
Traffic and Tickets simulations that make those relationships visible through interaction.
Design decision
I made the relationships manipulable so someone can compare a change with its downstream effects.
Where it stands
The simulations are teaching models with simplified rules, rather than calibrated forecasts of a real transport or support system.
Take a look
Inspect the Traffic simulation screenshot, including the controls for changing demand and transport choices.

View the visual reference ↗

AI literacy learning resource · Local learning prototype · interface example available

Tiny App Clinic

Learn through a small app and the practical friction that comes with building it.

Topics: Learning & enablement

Available: Screenshots

Problem
Learners and teachers need ways to discuss why a small build works, fails, or needs revision.
Built
A student-facing build guide with a teacher-facing layer for diagnosing learning and workflow friction.
Where it stands
The example demonstrates an authored local ruleset, not a live model response or a deployed classroom intervention.
Take a look
Inspect a scoped repair plan in the original student-facing interface.

View the visual reference ↗

Teacher workflow exploration · Local teaching workflow · sample screenshot available

Reteach Sketchpad

Help a teacher examine learner responses and decide what to revisit.

Topics: Learning & enablement

Available: Screenshots

Problem
Instructional support needs to keep teacher judgment central and handle learner evidence carefully.
Built
A teacher-facing browser build with editable evidence, a suggested reteach move, revision controls, and a copyable lesson plan.
Design decision
I kept the recommendation beside the evidence and teacher judgment prompts, without scoring or grouping individual students.
Where it stands
The current analysis is deterministic and uses fictional responses. Classroom use and learning outcomes have not been established.
Take a look
Inspect the sample input and plan together, including what the tool can and cannot infer.

View the visual reference ↗

College planning exploration · Local build · sample dashboard available

ApplyClear

Bring structure to college-planning decisions.

Topics: Learning & enablement

Available: Screenshots

Problem
College choices involve several competing priorities that are hard to compare informally.
Built
A higher-education decision-support exploration. This description does not offer admissions advice or a live service.
Where it stands
This capture uses the built-in sample student profile. Admissions, cost, and scholarship figures are illustrative and must be verified independently.
Take a look
Inspect the original sample dashboard and its student planning assumptions.

View the visual reference ↗

Experiments

Experiments

Ideas in progress, shown honestly as prototypes and explorations.

Experimental prototype · Local build · screenshot available

Unfinished University

A fictional learning institution where proposals, consent, and evidence are part of the experience.

Topics: Creative explorations

Available: Screenshots

Problem
Learning experiences can rush to a polished answer before exploring uncertainty.
Built
An interactive prototype for exploring proposals, learning experiences, and decisions inside a fictional institution.
Where it stands
The website and learning workflow are prototypes. The sample form does not establish a live institution or participant outcomes.
Take a look
Inspect the original interface capture.

View the visual reference ↗

Experimental prototype · Local build · screenshot available

Bureau of Uninvented Work

A fictional casework system for exploring jobs and practices that do not yet exist.

Topics: Creative explorations

Available: Screenshots

Problem
New kinds of work are often named before their responsibilities and evidence are understood.
Built
A casework prototype for taking an idea through intake, a field trial, and a final dossier.
Where it stands
The Bureau is an experimental casework concept. The capture shows a built interface, not demonstrated field-trial outcomes.
Take a look
Inspect the original interface capture.

View the visual reference ↗

Consent demonstration · Local build · screenshot available

The Contextual Self

Show how a service could receive one needed fact rather than a complete identity profile.

Topics: Agents & permissions

Available: Screenshots

Problem
A request for identity information can ask for much more than its stated purpose requires.
Built
A fictional age-verification demonstration with limited disclosure, a consent history, refusal, and revocation. Authorization and interpretation are deterministic simulations.
Where it stands
Identity facts and authority matching are deterministic fixtures. This is a consent-flow demonstration, not a deployed identity or age-verification service.
Take a look
Inspect the original interface capture.

View the visual reference ↗

Simulation exploration · Local build · screenshot available

Tiny Living Universe

Explore a small world through visual atmosphere and changing system states.

Topics: Creative explorations

Available: Screenshots

Problem
Emergent behavior is difficult to communicate through a static explanation.
Built
A creative browser simulation exploring world logic and interactive visual systems.
Where it stands
Population, culture, and events follow simplified simulation rules. They are not predictions about real communities.
Take a look
Inspect the original interface capture.

View the visual reference ↗

Narrative game exploration · Local build · screenshot available

Trial of the Monster

Use a browser game to explore judgment and interactive storytelling.

Topics: Creative explorations

Available: Screenshots

Problem
A narrative can offer decisions that change how a player understands a situation.
Built
A local game exploration combining judgment mechanics, interactive narrative, and art-direction experiments.
Where it stands
This is an experimental narrative experience, not a therapeutic or clinical assessment.
Take a look
Inspect the original interface capture.

View the visual reference ↗

Satirical simulation · Local build · screenshot available

One More Demo

A playful founder simulation about the pressure to keep producing the next AI demo.

Topics: Creative explorations

Available: Screenshots

Problem
Interactive satire can make the tensions around technology development easier to examine.
Built
A browser exploration using 3D interaction and simulation mechanics to examine the AI boom.
Where it stands
This is an experimental browser satire. In-game money, customer activity, and results are simulation values.
Take a look
Inspect the original interface capture.

View the visual reference ↗

Collaborative game exploration · Local build · screenshot available

The World Between Us

Explore a fictional world where one player’s choices create consequences for others.

Topics: Creative explorations

Available: Screenshots

Problem
Shared-world stories need ways for separate perspectives and decisions to affect one another.
Built
The Lantern Assembly, a fixed eight-role local fixture with linked quests and a game-master review console. It is not an online multiplayer service.
Where it stands
Roles and events are fictional local fixtures. The interface does not establish an online multiplayer service.
Take a look
Inspect the original interface capture.

View the visual reference ↗

Mobile app exploration · Local mobile exploration · character artwork available

LowKey Move

Turn starting a task into a small quest with a companion character.

Topics: Creative explorations

Available: Visual preview

Problem
A large or unappealing task can be difficult to begin.
Built
A local mobile app exploration with task-starting interactions and a layered animated mascot. User testing and public distribution are not established.
Where it stands
The references establish the character direction. User testing, task-starting outcomes, and public app distribution have not been established.
Take a look
Inspect the character artwork.

View the visual reference ↗

Research & frameworks

Research & frameworks

Explorations of evidence, governance, and human decisions in AI workflows.

Workflows & operations

Discover, plan, and coordinate everyday work. · Resources

Workflow planning toolkit · Public source package

Stack Engine

Turn an operational goal into an inspectable AI workflow plan.

Topics: Workflows & operations

Available: Source code

Problem
An idea to automate something still needs a decision about what to build and what to keep manual.
Built
A workflow architecture tool with diagnosis, stack suggestions, implementation steps, deterministic scoring, and prompt packs.
Where it stands
Its scores and verdicts are planning recommendations, not measured operational outcomes.
Take a look
Browse the public repository, documentation.

Explore the repository ↗

Agents & permissions

Route agent work and keep access, authority, and human decisions visible. · Resources

Authority classification protocol · Public source package · working paper available

Agent Authority Protocol

Classify what an AI agent may do, what it can affect, and whether it may delegate.

Topics: Agents & permissions

Available: Download · Source code · Sample files

Problem
An agent’s technical access does not describe the authority people have given it.
Built
An Authority–Reach–Delegation model, class definitions, risk scorecard, example dossiers, and a working paper.
Where it stands
This is an open classification proposal. It complements access controls and human governance; it does not enforce permissions at runtime.
Take a look
Browse the public repository, documentation and examples.

Explore the repository ↗

Read the working paper ↗

Inspect the public examples ↗

Agent context toolkit · Public source package

Context Engine

Choose and package repository context within an explicit token budget.

Topics: Agents & permissions · Releases & handoffs

Available: Source code · Sample files

Problem
Giving an agent an entire repository can bury important files and hide exclusions.
Built
A context scanner with file classification, token accounting, trimming, exclusions, and inspectable manifests.
Where it stands
A manifest describes included context and budget decisions. It does not guarantee an agent will understand or use the context correctly.
Take a look
Browse the public repository, documentation and examples.

Explore the repository ↗

Inspect the public examples ↗

Agent work packet toolkit · Public source package

Handoff Engine

Turn an objective and prepared context into a bounded agent work order.

Topics: Agents & permissions · Releases & handoffs

Available: Source code · Sample files

Problem
Loose instructions can omit constraints, allowed files, or completion checks.
Built
A Python CLI that produces objectives, context, constraints, acceptance checks, and proof cases in handoff packets.
Where it stands
The packet documents a work order. It does not execute the work or establish that the agent followed it.
Take a look
Browse the public repository, documentation and examples.

Explore the repository ↗

Inspect the public examples ↗

Local credential boundary · Public source package

Zero-Env Proxy

Let approved automation workers call configured services without receiving provider credentials.

Topics: Agents & permissions

Available: Source code · Sample files

Problem
A worker may need to use a service without carrying its secret directly.
Built
A local proxy with service allowlists, file enrollment, hash checks, and a key-free mock-provider demo.
Where it stands
The boundary is local and file-based. It is not an organization-wide vault or cryptographic process identity system.
Take a look
Browse the public repository, documentation and examples.

Explore the repository ↗

Inspect the public examples ↗

Evidence & evaluation

Examine claims, test results, uncertainty, and supporting records. · Resources

Research workflow exploration · Public source package

Contested Evidence Autoresearch

Organize unresolved questions while preserving conflicting evidence.

Topics: Evidence & evaluation

Available: Source code · Sample files

Problem
A fluent AI answer can hide uncertainty when the underlying sources disagree.
Built
A Markdown field kit with research questions, competing theory maps, evidence cards, study autopsies, cycle reports, and example question backlogs.
Where it stands
The kit supports structured human research under uncertainty. It is not an autonomous research engine or a substitute for formal evidence review.
Take a look
Browse the public source package, documentation, and available examples.

Explore the public repository ↗

Inspect the public examples ↗

Evidence review tool · Public source package

u27-scan

Require source evidence and an explicit failure condition before producing a finding.

Topics: Evidence & evaluation

Available: Source code

Problem
A report can look credible even when a finding lacks support.
Built
A command-line and static-report workflow that enforces evidence requirements before rendering findings.
Where it stands
The supported lane validates human- or agent-supplied candidates against exact source quotations and required fields. Live model extraction is experimental and needs the user’s own credentials.
Take a look
Browse the public source package, documentation, and available examples.

Explore the public repository ↗

Claim review toolkit · Public source package

Proofwashing Suite

Examine whether a claim is supported by the evidence attached to it.

Topics: Evidence & evaluation

Available: Source code · Sample files

Problem
Confident claims can outrun the records available to support them.
Built
An orientation package for five local claim-review tools: Humility Engine, Evidence Floor, Proof Decay, Caveat Drop, and Claim Drift.
Where it stands
The suite helps a human choose and sequence review tools. It does not verify truth or automatically approve publication.
Take a look
Browse the public source package, documentation, and available examples.

Explore the public repository ↗

Inspect the public examples ↗

Claim review utility · Public source package

Caveat Drop

Spot limitations that disappear between evidence notes and public copy.

Topics: Evidence & evaluation

Available: Source code · Sample files

Problem
A narrowly supported result can sound much broader when its caveats are removed.
Built
A local review tool that compares claims with structured caveat notes and proposes bounded wording.
Where it stands
Matching is heuristic. Findings and suggested wording require human review.
Take a look
Browse the public repository, documentation and examples.

Explore the repository ↗

Inspect the public examples ↗

Draft comparison utility · Public source package

Claim Drift

Review whether a revised claim became stronger without new evidence.

Topics: Evidence & evaluation

Available: Source code · Sample files

Problem
Editing can increase confidence or scope while the supporting record stays the same.
Built
A before-and-after draft comparison with evidence notes, risk flags, and a reviewer-facing report.
Where it stands
The comparison surfaces wording changes and possible evidence gaps; it does not determine whether either draft is true.
Take a look
Browse the public repository, documentation and examples.

Explore the repository ↗

Inspect the public examples ↗

Evidence policy utility · Public source package

Evidence Floor

Check whether a claim has the kind of evidence its declared policy requires.

Topics: Evidence & evaluation

Available: Source code · Sample files

Problem
Different claims need different levels of support.
Built
A local tool that compares claims, evidence notes, and an evidence-floor policy in a review table.
Where it stands
Results depend on the supplied policy and notes. The tool supports human review, not certification.
Take a look
Browse the public repository, documentation and examples.

Explore the repository ↗

Inspect the public examples ↗

Evidence review utility · Public source package

Humility Engine

Examine where polished writing may outrun the evidence behind it.

Topics: Evidence & evaluation

Available: Source code · Sample files

Problem
Uncertain or narrowly supported information can become confident public language.
Built
A local draft-and-evidence review instrument with claim tables, scope flags, and suggested narrower wording.
Where it stands
This is a heuristic review aid. It does not corroborate external sources or establish truth.
Take a look
Browse the public repository, documentation and examples.

Explore the repository ↗

Inspect the public examples ↗

Evidence freshness utility · Public source package

Proof Decay

Review whether an older artifact still supports a current claim.

Topics: Evidence & evaluation

Available: Source code · Sample files

Problem
A past run can keep being cited after the system or its operating conditions change.
Built
A local comparison of current claims with structured proof notes, including stale-artifact and changed-condition flags.
Where it stands
Freshness and matching are heuristic. A review report does not validate the underlying evidence.
Take a look
Browse the public repository, documentation and examples.

Explore the repository ↗

Inspect the public examples ↗

Public claim review utility · Public source package

Boundary Engine

Compare repository claims with recorded proof before publishing.

Topics: Evidence & evaluation

Available: Source code · Sample files

Problem
Public wording can imply more readiness than the recorded checks support.
Built
A Python CLI that scans Markdown claims against a proof ledger and writes a boundary register and JSON report.
Where it stands
It checks supplied claims against supplied proof records. It does not independently verify truth or rewrite documents.
Take a look
Browse the public repository, documentation and examples.

Explore the repository ↗

Inspect the public examples ↗

Evaluation toolkit · Public source package

Eval Bench

Run declared evaluation cases and retain their results.

Topics: Evidence & evaluation

Available: Source code · Sample files

Problem
Acceptance criteria need inspectable execution records before they become completion claims.
Built
A deterministic evaluation runner producing result JSON, evidence files, a report, and proof-case definitions.
Where it stands
Passing declared cases supports those checks, not broad model quality or production readiness.
Take a look
Browse the public repository, documentation and examples.

Explore the repository ↗

Inspect the public examples ↗

Execution evidence toolkit · Public source package

Proof Ledger

Record commands, exit results, and supporting files as durable proof artifacts.

Topics: Evidence & evaluation

Available: Source code · Sample files

Problem
A remembered terminal result is difficult for another reviewer to inspect later.
Built
A local runner and ledger that retain command evidence, failed runs, and reviewer-facing proof packets.
Where it stands
The ledger records local executions. It does not independently establish production behavior or certify claims.
Take a look
Browse the public repository, documentation and examples.

Explore the repository ↗

Inspect the public examples ↗

Learning & enablement

Build practical skills and help people make changes in their work. · Resources

Implementation resources · Public source package

Unit27 Field Kit Suite

Practical patterns for defining AI workflow boundaries and readiness.

Topics: Learning & enablement · Workflows & operations

Available: Source code

Problem
An implementation needs usable working guidance as well as an idea for what AI might do.
Built
An eight-step public tooling sequence: Stack Engine, Context Engine, Knowledge Readiness, Handoff Engine, Eval Bench, Proof Ledger, Boundary Engine, and u27-check.
Where it stands
The suite is an orientation and operating guide. Individual tools retain their own installation requirements and evidence limits.
Take a look
Browse the public source package, documentation, and available examples.

Explore the public repository ↗

Google Docs helper · Public source package

Open APA Desk

Prepare paper sections, references, and citations in Google Docs.

Topics: Learning & enablement

Available: Source code

Problem
Students need a practical route from source details to readable citations and a references section.
Built
An Apps Script alpha for common APA 7 reference types, DOI metadata lookup, citation insertion, and preparing a clean document copy.
Where it stands
The repository is public. The alpha is not approved for installation through Google Workspace Marketplace and does not cover every APA edge case.
Take a look
Browse the public repository, documentation.

Explore the repository ↗

Releases & handoffs

Package, inspect, and hand over software and its supporting files. · Resources

Artifact packaging utility · Public source package

Artifact Manifest

Package supplied artifacts with hashes, declared scope, and limitations.

Topics: Releases & handoffs

Available: Source code · Sample files

Problem
A screenshot or test output needs context about what it actually supports.
Built
A local Rust tool that writes JSON and Markdown evidence packets, including file hashes and risk flags.
Where it stands
Hashes establish file identity. The packet records supplied artifacts and declared boundaries; it does not verify a claim’s truth.
Take a look
Browse the public repository, documentation and examples.

Explore the repository ↗

Inspect the public examples ↗

Release review workspace · Public source package

Release Desk

Bring release claims, artifacts, boundaries, and approval gates into one review surface.

Topics: Releases & handoffs

Available: Source code

Problem
A release decision needs more than a healthy build or an unfinished checklist.
Built
A small Rails application with synthetic seed data and editable local release, claim, artifact, and gate records.
Where it stands
This is a local human-review workspace. It does not publish releases or certify that a repository is safe.
Take a look
Browse the public repository, documentation.

Explore the repository ↗

Repository review utility · Public source package

Source Index

Map the files and public claims available for a repository release review.

Topics: Releases & handoffs

Available: Source code · Sample files

Problem
Missing documentation, examples, or boundary notes can be difficult to notice in a large repository.
Built
A Go CLI that writes Markdown and JSON maps of release surfaces, boundary language, and risky phrases.
Where it stands
It identifies visible files and wording. It does not audit application behavior or certify release readiness.
Take a look
Browse the public repository, documentation and examples.

Explore the repository ↗

Inspect the public examples ↗

Launch preflight utility · Public source package

u27-check

Check a local build’s entry point, homepage, and primary action.

Topics: Releases & handoffs

Available: Source code

Problem
A project can build successfully while its first visitor path is broken.
Built
A deterministic preflight tool for build execution, runtime load, entry detection, and primary-action target checks.
Where it stands
Checks cover the first local user path. They do not establish deeper product quality, backend correctness, or hosted availability.
Take a look
Browse the public repository, documentation.

Explore the repository ↗

Evidence & evaluation

Examine claims, test results, uncertainty, and supporting records. · Images

Human review demonstration · Local build · screenshot available

Verdelta

Explore how evidence and human decision points can be built into an AI workflow.

Topics: Evidence & evaluation

Available: Screenshots

Problem
A model’s proposal needs to remain distinct from permission to act.
Built
A local synthetic-data prototype for human review and bounded AI authority. It is not a production compliance or authorization system.
Where it stands
The prototype uses synthetic data. Redacted fields are intentionally unavailable; the image does not establish production compliance or authorization.
Take a look
Inspect the original interface capture.

View the visual reference ↗

Websites & practices

Websites & practices

The public homes for my implementation, web design, and independent research work.

AI implementation practice · Public website

Gig City AI

Practical AI workflows, tools, and enablement resources.

Topics: Websites & portfolios

Available: Live site

Problem
Organizations need a way to connect AI capabilities with everyday operating needs.
Built
My AI implementation practice and its website, bringing together workflow work, learning resources, and practical tools.
Take a look
Explore the practice and its work on the public website.

Visit the website ↗

Web design practice · Public website

Gig City Web Design

Web work for small businesses, with a clear route to the information visitors need.

Topics: Websites & portfolios

Available: Live site

Problem
A small-business website needs to make the business understandable and easy to contact.
Built
A separate website and business surface for web design work.
Take a look
Explore the practice and its work on the public website.

Visit the website ↗

Independent research initiative · Public website

Unit27 Research

Frameworks and experiments about organizational design, governance, and human judgment.

Topics: Websites & portfolios

Available: Live site · Source code

Problem
New AI workflows raise questions about evidence, responsibility, and the relationship between people and systems.
Built
My independent research and intellectual-property initiative, with a public website for its work.
Take a look
Explore the practice and its work on the public website.

Visit the website ↗

Browse the public release index ↗

Related work

Gig City AI · Gig City Web Design · Unit27 Research