repo-standard/register/repos.json

{
  "$comment": [
    "Taxonomy register for the `open/` organisation (D4: central, one file).",
    "The class is READ OFF the catalog and the remotes — it is structural, not a judgement.",
    "Refresh the `repos` name set against ground truth with:",
    "  node scripts/repo-standard-check.mjs --refresh",
    "which enumerates /api/v1/orgs/open/repos (ONE call — the listing carries",
    "description and topics too; per-repo fetching trips the rate limiter).",
    "Enumerate, never glob: 12 of these sit at depth 2 locally, one has a",
    "basename that differs from its repo name, and `.profile` is hidden."
  ],
  "org": "open",
  "forge": "https://git.fromaitochitta.com",
  "marketplace": {
    "name": "ktg-plugin-marketplace",
    "url": "https://git.fromaitochitta.com/open/ktg-plugin-marketplace.git"
  },
  "repos": {
    "llm-security": "plugin",
    "config-audit": "plugin",
    "voyage": "plugin",
    "linkedin-studio": "plugin",
    "graceful-handoff": "plugin",
    "ai-psychosis": "plugin",
    "ms-ai-architect": "plugin",
    "okr": "plugin",
    "human-friendly-style": "plugin",
    "claude-design": "plugin",
    "repo-mailbox": "plugin",
    "repo-standard": "plugin",
    "ktg-plugin-marketplace": "catalog",
    "playground-design-system": "shared-asset",
    "portfolio-optimiser-commons": "shared-asset",
    "llm-security-commons": "shared-asset",
    ".profile": "org-profile",
    "portfolio-optimiser": "standalone",
    "portfolio-optimiser-claude": "standalone",
    "llm-ingestion-pipeline-security": "standalone",
    "llm-ingestion-okf": "standalone"
  },

  "$comment_non_repos": [
    "Names that LOOK like repo names and are not. These exist so that",
    "'no match' and 'match on something that is not a repo' are DIFFERENT",
    "outcomes — if they share an outcome, the loss goes silent, which is the",
    "defect class this whole standard exists to catch.",
    "Each entry is a measured false positive, not a guess."
  ],
  "non_repos": {
    "coord": "Retired repo name, deliberately still alive in prose: the CLI (coord-send), the mailbox root (~/.claude/coord/) and CLAUDE_COORD_DIR kept it — they are the transport protocol, not the product. The repo has been `repo-mailbox` since v0.3.0.",
    "_broadcast": "Reserved engine namespace in the coord mailbox (`~/.claude/coord/_broadcast/`). Occupies a repo-shaped PATH position; `_` prefixed names are refused as repo identities.",
    "llm-ingestion-guard": "Package name published by `llm-ingestion-pipeline-security`. A package, not a repo.",
    "claude-code-llm-security": "Pre-split name of `llm-security`. This one IS dead — the org's only rename produced every dead reference we found. Listed so the finding names the successor instead of just failing."
  },

  "$comment_classes": [
    "Per class: required files, required README headings, and the install form.",
    "A flat standard across all classes would demand a ROADMAP from a 5-line",
    "profile. ROADMAP is deliberately absent everywhere: it is 0/18 today and is",
    "drafted from STATE by a human. A gate that fails every repo teaches people",
    "to switch the gate off.",
    "",
    "CONTRIBUTING.md, CODE_OF_CONDUCT.md and MAINTAINERS.md are deliberately NOT",
    "required anywhere. The maintainer works alone and the catalog's published",
    "stance already says so — 'solo-maintained, fork-and-own; issues welcome as",
    "signals, pull requests not accepted'. Contributor-facing documentation for a",
    "project that accepts no contributors is theatre, and a CODE_OF_CONDUCT with",
    "an unattended placeholder address is worse than none: it is a visible",
    "unfinished template. This is NOT a rule against having them — files already",
    "present are a separate cleanup decision, not a gate finding.",
    "",
    "Consumer-facing documentation is unaffected by working alone, and that is",
    "the whole distinction: SECURITY.md, LICENSE, CHANGELOG, non-goals and honest",
    "limitations exist for the reader, not for a contributor."
  ],
  "classes": {
    "plugin": {
      "required_files": ["README.md", "LICENSE", "CHANGELOG.md", ".claude-plugin/plugin.json"],
      "required_headings": ["## Install", "## Non-goals", "## Changelog"],
      "install": "plugin"
    },
    "catalog": {
      "required_files": ["README.md", "LICENSE", "GOVERNANCE.md", "CONVENTIONS.md", ".claude-plugin/marketplace.json"],
      "required_headings": ["## Install", "## Non-goals"],
      "install": "catalog"
    },
    "shared-asset": {
      "required_files": ["README.md", "LICENSE"],
      "required_headings": ["## Non-goals"],
      "install": "vendor"
    },
    "org-profile": {
      "required_files": ["README.md"],
      "required_headings": [],
      "install": "none"
    },
    "standalone": {
      "required_files": ["README.md", "LICENSE"],
      "required_headings": ["## Install", "## Non-goals"],
      "install": "package"
    }
  },

  "$comment_traits": [
    "A SECOND axis, orthogonal to class. Class is structural (read off the",
    "catalog and the remotes); a trait is about what the code DOES, which no",
    "remote can tell you. `security` attaches the obligations a tool acquires by",
    "handling untrusted input: a real disclosure channel, and limitations stated",
    "with their mechanism.",
    "",
    "PROPOSED BY THE GATE'S AUTHOR, NOT MEASURED — the operator owns this list.",
    "Marking a repo `security` decides that it owes a SECURITY.md, so adding or",
    "removing a name here is a judgement, not a reading. These two were picked",
    "because both process untrusted input as their stated purpose."
  ],
  "traits": {
    "llm-security": ["security"],
    "llm-ingestion-pipeline-security": ["security"]
  },
  "trait_requirements": {
    "security": {
      "required_files": ["SECURITY.md"],
      "required_headings": ["## Known limitations"]
    }
  },

  "$comment_locales": [
    "A THIRD axis, orthogonal to both class and trait. Class is structural, a",
    "trait is what the code DOES — this is who the code is FOR, which is the",
    "standard's own stated principle: who the reader is decides what is",
    "required. English is the default and is not listed. A repo aimed ONLY at a",
    "Norwegian readership is `nb`, and is then WRONG in English, not right.",
    "",
    "PROPOSED BY THE OPERATOR, NOT MEASURED — like `traits`, this list is a",
    "judgement and no remote can report it. Both entries below were named by",
    "the operator on 2026-08-04 as Norway-only in their audience.",
    "",
    "Detection is a stopword-frequency comparison over prose with code stripped.",
    "It answers WHICH language dominates, never whether the prose is any good."
  ],
  "locales": {
    "ms-ai-architect": "nb",
    "okr": "nb"
  },

  "description_max_codepoints": 180,
  "$comment_length": [
    "180 codepoints, not bytes and not UTF-16 units. The same string measures 248",
    "/ 249 / 253 across the three yardsticks (graceful-handoff: `👉` is astral).",
    "An em-dash costs 3 bytes but 1 codepoint AND 1 UTF-16 unit, so it exposes",
    "only the outer layer and hides the inner one. JS-based tooling reads one",
    "higher per astral character. Upper bound: 207 nearly filled the card's text",
    "field; 220 is untested and may overflow."
  ]
}