{"id":51505063,"url":"https://github.com/watadani-byte/versionable-ai-assets","last_synced_at":"2026-07-07T23:02:12.482Z","repository":{"id":349123605,"uuid":"1201136812","full_name":"watadani-byte/versionable-ai-assets","owner":"watadani-byte","description":"A practical framework for treating AI operational inputs — prompts, images, constraints, and reference materials — as versionable, auditable assets.","archived":false,"fork":false,"pushed_at":"2026-04-04T10:27:04.000Z","size":49,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-04-04T11:31:11.877Z","etag":null,"topics":["ai-governance","ai-operations","ai-workflow","asset-management","operational-assets","prompt-engineering","version-control","versionable"],"latest_commit_sha":null,"homepage":"https://watadani-byte.github.io/character-identity-protocol/","language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/watadani-byte.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-04-04T09:03:00.000Z","updated_at":"2026-04-04T10:27:08.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/watadani-byte/versionable-ai-assets","commit_stats":null,"previous_names":["watadani-byte/versionable-ai-assets"],"tags_count":1,"template":false,"template_full_name":null,"purl":"pkg:github/watadani-byte/versionable-ai-assets","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/watadani-byte%2Fversionable-ai-assets","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/watadani-byte%2Fversionable-ai-assets/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/watadani-byte%2Fversionable-ai-assets/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/watadani-byte%2Fversionable-ai-assets/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/watadani-byte","download_url":"https://codeload.github.com/watadani-byte/versionable-ai-assets/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/watadani-byte%2Fversionable-ai-assets/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":35245324,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-07-07T02:00:07.222Z","response_time":90,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai-governance","ai-operations","ai-workflow","asset-management","operational-assets","prompt-engineering","version-control","versionable"],"created_at":"2026-07-07T23:02:11.915Z","updated_at":"2026-07-07T23:02:12.475Z","avatar_url":"https://github.com/watadani-byte.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# Versionable AI Assets\n\n*A working framework for managing prompts, images, manuals, and constraints as versioned operational assets.*\n\n-----\n\n## Who This Is For\n\nThis repository is for teams and practitioners who need AI workflows to be reusable, reviewable, and stable across time — especially when prompts, images, manuals, and constraints are reused across sessions, operators, or projects.\n\n-----\n\n## What This Is Not\n\nThis is not a generic prompt library, a model benchmark repository, or a collection of one-off generation tricks. It is a framework for treating operational inputs as managed assets.\n\n-----\n\n## What This Repository Is\n\nThis repository proposes a practical operational framework for AI asset management:\n\n\u003e Prompts, images, manuals, constraints, and reference materials used in AI workflows\n\u003e should be treated as versionable operational assets —\n\u003e not as informal, disposable inputs.\n\nThis is not presented as a new theory.\nIt is a practical application of established software and documentation practices\nto AI operational inputs.\n\n-----\n\n## Start Here\n\n- Read `docs/concept.md`\n- Then read `docs/versioning_policy.md`\n- Then inspect the templates in `docs/templates/`\n\n-----\n\n## The Problem\n\nAI workflows are often managed informally.\n\nPrompts are edited in place.\nImages are stored without version labels.\nReference documents accumulate without structure.\nNobody records what changed, when, or why.\n\nThis works for small, one-off tasks.\nIt breaks down when:\n\n- Workflows run across multiple sessions or operators\n- Outputs must be reproducible or auditable\n- Assets need to be shared, updated, or rolled back\n- Quality regressions need to be diagnosed\n\nIn many cases, the operational bottleneck is not the model itself, but the lack of structured asset management around it.\n\n-----\n\n## The Proposal\n\nTreat AI operational inputs as **versionable assets**.\n\nThat means:\n\n- each asset has an explicit role\n- each asset has a version\n- changes are tracked with reasons\n- rollback is always possible\n- active assets are separated from archived or deprecated ones\n- research materials are separated from execution materials\n\nThis is standard practice in software and documentation workflows.\nIt is still unevenly applied in AI operations.\n\n-----\n\n## Core Concept: The Versionable AI Asset\n\nA **Versionable AI Asset** is any unit of input material used in an AI workflow that is:\n\n- **meaningful** — it has a defined operational role\n- **bounded** — it has clear scope and edges\n- **reusable** — it can function across multiple sessions\n- **replaceable** — it can be updated modularly without rebuilding everything\n- **auditable** — its history is traceable\n\nAssets may include:\n\n|Asset Type          |Examples                                               |\n|--------------------|-------------------------------------------------------|\n|Identity definitions|YAML character sheets, persona specifications          |\n|Anchor references   |Validated images, structural reference sets            |\n|Behavioral profiles |Tone guides, register definitions, role constraints    |\n|Operational manuals |Workflow instructions, session management rules        |\n|Constraint sets     |Acceptance criteria, quality gates, hard limits        |\n|Lexicons            |Controlled vocabulary for modifications                |\n|Research materials  |Drift documentation, failure logs, comparative analysis|\n\nNot every AI input needs to be managed this way.\nSimple, one-off prompts do not require version control.\n\nThe framework is intended for inputs that are **reused, updated, shared, or relied upon over time**.\n\n-----\n\n## Key Distinctions\n\nThis framework rests on several distinctions that are easy to blur in practice.\n\n### Execution assets vs. research assets\n\n**Execution assets** are live operational materials — narrow, validated, and actively used.\n\n**Research assets** are exploratory materials — broader, messier, and not yet validated for production use.\n\nThese should not be stored in the same location or mixed into the same workflows.\n\n### Active vs. archived vs. deprecated\n\n|Status        |Meaning                                                  |\n|--------------|---------------------------------------------------------|\n|`active`      |Currently in operational use                             |\n|`archive`     |Preserved for rollback or comparison — not for active use|\n|`deprecated`  |No longer recommended — retained for traceability only   |\n|`experimental`|Under evaluation — not yet validated for production      |\n\n### YAML as SSOT\n\nFor structured definitions such as identity, behavior, and constraints, YAML can serve as the **Single Source of Truth**.\n\nEnglish-language prompts are a delivery format — a projection of the YAML definition.\nWhen generation quality degrades, one robust response is to return to the YAML source\nrather than continuing to patch the delivery prompt in place.\n\n-----\n\n## Versioning Model\n\nAll assets should follow a three-part version scheme: `MAJOR.MINOR.PATCH`\n\n|Change type|When to use                              |Example          |\n|-----------|-----------------------------------------|-----------------|\n|Major      |Structural redesign, core logic change   |`v1.0.0 → v2.0.0`|\n|Minor      |Meaningful improvement, new content added|`v0.1.0 → v0.2.0`|\n|Patch      |Small corrections, formatting, typos     |`v0.1.0 → v0.1.1`|\n\nEvery version change should include a brief revision note explaining what changed and why.\n\n-----\n\n## Repository Structure\n\n```\nversionable-ai-assets/\n├── README.md\n├── docs/\n│   ├── concept.md              ← Core concept document\n│   ├── versioning_policy.md    ← Versioning rules and rationale\n│   ├── asset_taxonomy.md       ← Asset types and their roles\n│   └── templates/\n│       ├── asset_manifest.yaml ← Template: asset metadata\n│       ├── identity_core.yaml  ← Template: identity definition\n│       └── acceptance_criteria.md ← Template: quality criteria\n└── examples/\n    └── (to be added)\n```\n\n-----\n\n## Related Projects\n\n- [character-identity-protocol](https://github.com/watadani-byte/character-identity-protocol) — A governance framework for character identity continuity in generative AI. CIP and related PAL work are downstream examples of the versionable asset approach applied to a specific domain.\n\n-----\n\n## Status\n\nThis repository is in early development.\nThe core documents are working drafts.\nFeedback and discussion are welcome via Issues.\n\n-----\n\n## License\n\nThis repository is currently under a draft custom license.\nSome materials may later be separated under different terms.\nPlease do not reuse or redistribute without prior permission.\nDetailed licensing terms will be finalized in a later revision of this repository.\n\n© 2026 Hitoshi Watadani","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fwatadani-byte%2Fversionable-ai-assets","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fwatadani-byte%2Fversionable-ai-assets","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fwatadani-byte%2Fversionable-ai-assets/lists"}