Skip to content

Cut the model-read ecosystem specifics from code-tidying #16756

Open
5 of 5 issues completed
Open

Description

@kyle-sexton

Parent

Refs #15703
Source: #15730 (blind classification under rules revision 5, #15730 comment 6950077480), cut item 7 of 47 in its filing order.

Depends on

Blocks #15732 (the portability gate widens only after every cut lands).

Outcome

code-tidying ships no model-read tool, library or command choice outside ADR 0101's named exceptions: the 25 rows below are cut (R11 21, R4 3, R8 1), each R8 row unless its targeted case fails without it.

Demo: on main after the merge, each listed row's cited text is gone or replaced by generic text at a routing site, and the pull request body lists every row id with what happened to it.

Row count (bulk-mutation standard): 25 rows. The pull request states this number and lists each row.

Rows to cut

Paths are repository-relative; line numbers are as read at 5b66a7fa and may have moved. Kind is the inventory kind (K2 detection table, K3 bundled value, K4 tool guidance, K5 tool pointer, K7 example).

Row Rule Kind Cited at Fact the classifier matched Notes
F0092 R11 K4 plugins/code-tidying/skills/audit-dead-code/context/adjudication.md:7-32 The evidence-pattern table is model-read per-language adjudication guidance (MediatR, Flask, click, pytest, reflection idioms) that no earlier rule holds.
F0096 R8 K4 plugins/code-tidying/skills/remove-comments/reference/tooling.md:38-59 The dated measurements of outside tools (tree-sitter-language-pack failing behind a proxy, tree-sitter-powershell parse errors, tree-sitter-bash base-N rejection) contradict the reasonable default of using those tools.
F0097 R11 K5 plugins/code-tidying/skills/remove-comments/reference/tooling.md:61-82 The C# compiler's CS1572/CS1573, Ruff DOC, eslint-plugin-jsdoc and Vale are named as the things to use for prose and contradiction checks, and no kept adapter drives them. Two readings (other: R6). Reader's note: F0097: lines 61-75 list the skill's own scripts (R6), but the row's ecosystem content is the tool pointers at 76-83, so I placed it under R11.
F0100 R4 K3 plugins/code-tidying/skills/remove-comments/reference/safety.md:110-114 The test-net step falls back to an ecosystem default runner (dotnet test, npm test, pytest, go test ./..., bats), a value that picks a tool for the consumer.
F0101 R11 K4 plugins/code-tidying/skills/remove-comments/reference/safety.md:128-152 The exempt-surface list is model-applied per-language guidance (docstrings, C# XML docs, JSDoc/TSDoc, GoDoc, PowerShell help, the Python underscore rule); the text says the exemption is the list's job, not the proof's. Two readings (other: GONE). Reader's note: F0102: answered GONE because the cited file no longer holds the list. If a move to another file of the same skill (reference/keep-list.md:7) counts as still present, the answer is R11, as for F0101.
F0104 R11 K5 plugins/code-tidying/skills/remove-comments/reference/triage.md:146-150 Vale is named as the optional prose linter to use where a repository runs it, a tool pointer no kept adapter drives.
F0105 R11 K4 plugins/code-tidying/skills/remove-comments/reference/triage.md:158-167 The doc-comment section of triage.md (docstrings, C# XML docs, JSDoc/TSDoc exempt) is still present and is model-read K4 guidance no earlier rule holds; remove-comments' description names no language.
F0108 R11 K7 plugins/code-tidying/skills/tidy/SKILL.md:93, :176 The node --test ad hoc scope example (now SKILL.md:103) and the shellcheck/shfmt commit-title example (now :186) are K7 examples in tidy, whose description names no language.
F0110 R11 K3 plugins/code-tidying/skills/tidy/lanes/shell-tooling.md:5-16 The bundled shell-tooling Scope globs (tools/**/*.sh, scripts/**/*.ps1) are a model-read K3 default that picks no tool and that no script executes.
F0111 R11 K4 plugins/code-tidying/skills/tidy/lanes/shell-tooling.md:31-49 The shell-tooling watch-for patterns (ShellCheck SC2086/SC2292/SC2230/SC2034, PowerShell idioms) are model-read K4 guidance, and tidy's description does not name shell as its scope.
F0113 R4 K3 plugins/code-tidying/skills/tidy/lanes/shell-tooling.md:60-79 The bundled shell-tooling verification commands pick shellcheck, shfmt and PSScriptAnalyzer as the consumer's check tools.
F0114 R11 K5 plugins/code-tidying/skills/tidy/lanes/shell-tooling.md:88-99 The Bash/PowerShell research sources (Vidar Holen/ShellCheck, Greg Wooledge/BashFAQ) are tool-specific pointers, not the basis of a language-neutral rule.
F0116 R11 K7 plugins/code-tidying/skills/tidy/reference/exclusions.md:25 Directory.Build.props and Directory.Packages.props are K7 "Typical examples" inside the consumer-declared protections, illustrations rather than the guard list itself.
F0117 R11 K7 plugins/code-tidying/skills/tidy/reference/exclusions.md:131-139 The ruff.toml entry in the overrides-file example (exclusions.md ~131-139) is a K7 example of a neutral file shape.
F0118 R11 K5 plugins/code-tidying/skills/tidy/reference/tidyings.md:23 tidyings.md:23 points to Roslyn, ruff F841 and biome as candidate surfacers, a model-read K5 tool pointer.
F0119 R11 K7 plugins/code-tidying/skills/tidy/reference/tidyings.md:55, :92, :176 tidyings.md:55 (C#/TS/Python/C), :92 (OrderValidationRules.cs) and :178 (npx markdownlint-cli2) are still present as K7 examples.
F0120 R11 K5 plugins/code-tidying/skills/tidy/lanes/docs-prose.md:30-36 The docs-prose watch-for P-1 names markdownlint-cli2 as what catches dead links, a model-read K5 tool pointer.
F0121 R4 K3 plugins/code-tidying/skills/tidy/lanes/docs-prose.md:49-55 The docs-prose verification command npx markdownlint-cli2 <changed .md files> picks markdownlint-cli2 as the consumer's lint tool.
F0123 R11 K7 plugins/code-tidying/skills/tidy/templates/dependency-root-lane.template.md:12, :24-25, :37-39 The dependency-root template's .Tests glob and e.g. dotnet build / npm run build / cargo build and biome comments are K7 placeholder examples.
F0124 R11 K7 plugins/code-tidying/skills/tidy/templates/polyglot-services-lane.template.md:10-11, :16-19, :34-35 The polyglot-services template's ruff F401/F841, biome and biome check && npm test / ruff check . && pytest comments are K7 examples.
F0125 R11 K7 plugins/code-tidying/skills/tidy/templates/apps-lane.template.md:11, :27 The apps template's <app-name>.Tests/** and **/*.razor are K7 examples.
F0126 R11 K7 plugins/code-tidying/skills/tidy/templates/host-wiring-lane.template.md:18 The host-wiring template's "one extension method" is C#-flavored K7 wording.
F0128 R11 K2 plugins/code-tidying/skills/batch-simplify/context/reference.md:11 reference.md:11 groups files by nearest *.csproj, package.json, pyproject.toml, Cargo.toml, a model-read K2 detection table.
F0129 R11 K7 plugins/code-tidying/skills/batch-simplify/SKILL.md:326 SKILL.md:326's "a .sh hook that invokes ruff" is a K7 example.
F0130 R11 K7 plugins/code-tidying/skills/tidy-diff/reference/generated-code-residue.md:42-89 The generated-code-residue detection cues are written in TS/JS syntax (as any, @ts-ignore, console.log) for neutral rules, K7 under a description that names no language.

Kept rows, not touched here (count per rule): R2 6, R3 5, R5 2 (D04, D05), R6 14, R7 2. 1 more rows were answered GONE (cited content no longer on main) and need nothing.

Steps

  • List every item with its rule: the pull request body repeats the table above with each row's outcome (cut, replaced at a routing site, or kept with its reason).
  • Back every R11 call (and each kept R5 row named below) with a search showing whether a script reads the file, for example git grep -n -F '<file name>' -- plugins scripts .github, quoted in the pull request body. A row whose file a script reads is not cut here: report it on Classify the remaining guidance rows blind and count the cuts per plugin #15730.
  • Name the cases that read the cut text: in the pull request body, name for each changed skill the existing eval case whose run reads the cut text, or say that none does. The lane names them; it runs none.
  • Whole page or part: in the pull request body, say for each changed file whether the cut deletes the whole page or part of it.
  • R4 rows: the bundled default goes; the choice is resolved from the consumer through the ecosystem-resolution contract (docs/contracts/ecosystem-resolution/). A shipped default that picks a library goes with this cut.
  • R8 rows: cut each like the other rows and mark it in the pull request body as pending its targeted case; the host session's result (below) decides whether it comes back.
  • Fence ratchet: lower the skill-stack-fences ceiling in .performance/ratchets.json for each of these files whose fences the cut removes: plugins/code-tidying/skills/audit-dead-code/context/adjudication.md.
  • References outside the plugin: these files on main name a cut file; update or remove each entry the cut makes stale: plugins/code-tidying/skills/audit-dead-code/context/adjudication.md in .performance/ratchets.json; plugins/code-tidying/skills/remove-comments/reference/tooling.md in scripts/docs-anchor-baseline.txt; plugins/code-tidying/skills/remove-comments/reference/safety.md in scripts/spoke-plugin-root-baseline.txt.
  • Open the pull request with Closes this issue in its body, then stop: the lane adds no label. A sweep adds do-not-merge and the hold line when the pull request opens; the evidence below is a host session's work.

After the pull request opens (host session, not the lane)

The lane opens the pull request and stops there. The pull request stays held with do-not-merge for the eval; a host session runs only the eval and posts its record on the pull request; the merge lane releases the hold when the record matches the head; this issue closes with the pull request.

  • Hold line: the hub's sweep (or a host session) adds do-not-merge to the pull request and posts, from a listed human's account, a comment with the line <!-- pr-hold reason=eval --> on a line of its own.
  • Targeted case per changed skill (this replaces Classify the remaining guidance rows blind and count the cuts per plugin #15730's "no more than 400 runs per plugin (the changed skills' cases on the opus and sonnet aliases, three runs per side, plus control sides on one model)"): one before-and-after case, on sonnet at 3 runs a side; opus runs only to confirm a sonnet loss; a new case gets a 1 to 3 run smoke first (owner decision of 2026-11-16 in the owner-decisions record, superseding the run counts in #15703 comment 6906800940). Select cases with --case, use --runs 3 and a --max-cost-usd ceiling, and start no batch after a usage-limit error.
  • Controls: each whole-page deletion carries a sensitivity-control result, its predicted outcome written in the pull request body before the first batch; each partial cut carries a detection-control result. No paired no-loss compare per cut (#15703 comment 6906800940).
  • On a LOSS reading: a trace read of each dropped case and an opus confirm; only a loss traced to the changed text counts (#16672 comment 6935228111).
  • R8 rows: the row's targeted case with and without the row. On a case drop or a mean fall of 0.15 or more without it, the host session restores the row on the pull request branch, annotated portability-ok: measured <case-id>; otherwise it stays cut. A deterministic grader states before the run in how many of n runs it must fail.
  • Record: the verdict and the control result go into the pull request body. When the run is VALID and shows no loss, the host session posts, from a listed human's account, a comment with the line <!-- eval-pair verdict=no-loss validity=VALID head=<40-hex head SHA> --> naming the pull request's current head, and keeps "do not merge" and do-not-merge out of that comment (docs/contracts/loop-lane/README.md, "Cross-lane PR hold"). A loss or a run that is not VALID leaves the hold in place; say so on the pull request.
  • Release: pr-merge's scheduled step "Release eval holds whose record matches the head" (.github/scripts/pr-merge/release-eval-hold.mjs) removes do-not-merge once that record is the newest since the label went on and names the current head, and no other hold stands. A push after the record needs a new record.
  • AC7 is met here: each whole-page deletion carries a sensitivity-control result and each partial cut a detection-control result, in the pull request body.

Decisions

  • D3: each plugin's cut is its own topic and pull request.
  • Interview Q32 (D4 of the Brief): bundled defaults that pick a tool, command or library are retired wherever they appear.
  • AC1's named exceptions are the kept classes of ADR 0101 (docs/adr/0101-resolve-ecosystem-specifics-from-the-consumer-repository.md, Decision and the 2026-11-17 amendment).
  • Bulk-mutation standard: count first, then act; the pull request states its row count.
  • The rules this item's rows fall under (rules revision 5):
    • R11 (shipped knowledge): everything else that is model-read K2, K3, K4, K5 or K7. Goes through the cutting route: the generic text replaces it only at a routing site where the skill still takes a language-specific step; anywhere else it is deleted with no replacement. An item kept only because a removed item consumes it goes in the same pull request.
    • R4 (consumer default): the value picks a tool or library for the consumer. Goes: no bundled default survives (ADR 0101 decision 2, "Ship no bundled default"); the choice is resolved from the consumer.
    • R8 (measured fact): it records a measurement or probe of an outside tool that contradicts a reasonable default, or pairs our rule id with an outside diagnostic. Moves into the record at drafting time when a record key can hold it; otherwise stays as one stamped row annotated portability-ok: measured <case-id> only if the row's targeted case shows a case drop or a mean fall of 0.15 or more without it, under the eval policy in force; a deterministic grader states before the run in how many of n runs it must fail; else goes.

Key interfaces

  • "The consumer's ecosystem (language, framework, tools): abstract now. Seam: the ecosystem-resolution contract and the consumer's own records."
  • Conventions followed: a cut ships with eval evidence (the skills-and-agents review rules).

Done when

  • AC1 (this plugin's share): code-tidying ships no model-read tool, library or command choice outside the named exceptions; each listed R11 and R4 row is cut, and each R8 row is cut and marked pending its targeted case.
  • The pull request body states the row count (25) and lists every row id with its rule and outcome.
  • Each R11 row's script-reader search is quoted in the pull request body.
  • The pull request body names, for each changed skill, the existing case that reads the cut text, or says none does.
  • The pull request body says, for each changed file, whether the cut deletes the whole page or part of it.
  • AC10: no new enforcement-placing skill is added here.

Out of scope

Activity

  1. added
    agent-readyFully specified and briefed; eligible for autonomous pickup from the frontier.
    on Nov 17, 2026
  2. added theissue type on Nov 17, 2026
  3. updated Work status ready and Priority medium on Nov 17, 2026
  4. github-actions commented on Nov 17, 2026

    @github-actions
    Contributor

    Size: Medium
    Complexity: Medium
    Uncertainty: Known
    Flags: none
    Blocked by: none
    Blast radius: about 20 markdown files under plugins/code-tidying (skills, lanes, templates, references), plus three baseline entries (.performance/ratchets.json, scripts/docs-anchor-baseline.txt, scripts/spoke-plugin-root-baseline.txt). Readers are the code-tidying skills' model runs.
    Reversible: yes; revert the pull request. The R8 rows can be restored on the branch if their targeted case shows a loss.
    Open questions: none
    I understood this as cutting 25 model-read ecosystem-specific rows (R11 21, R4 3, R8 1) from code-tidying, with the row count, per-row outcomes, script-reader searches, case names and whole-or-part notes stated in the pull request body. The issue states the outcome, steps and done-when exactly, so nothing is missing. The eval steps after the pull request opens are for a host session and are not part of this triage. I found no open duplicate and no merged pull request that already does this.

    Automated triage. A maintainer reviews it.

  5. melodic-automation-lanes commented on Nov 17, 2026

    @melodic-automation-lanes
    Contributor

    Not admitted to the issue lane yet: Size large: the split into sub-issues at intake is pending (owner Q93).

  6. added
    host-runNeeds an unattended run on a host; the host runner takes it, not a person or the issue lane.
    on Nov 17, 2026
  7. removed
    host-runNeeds an unattended run on a host; the host runner takes it, not a person or the issue lane.
    on Nov 17, 2026
  8. melodic-automation-lanes commented on Nov 17, 2026

    @melodic-automation-lanes
    Contributor

    The work-resolve-issue lane stopped on this issue (split-required): the issue's Size is too large for one run, and intake's split (owner Q93) files its sub-issues.
    Its Size is too large for one run: intake splits it into sub-issues (owner Q93).
    It set Work status to needs-triage, so the hourly triage sweep sends it back through intake triage, and it waits on no decision from a person.

    Run: https://github.com/melodic-software/claude-code-plugins/actions/runs/44146157029

  9. github-actions commented on Nov 17, 2026

    @github-actions
    Contributor

    Intake triage did not triage this issue after 2 dispatches, so the sweep no longer dispatches it and set its Work status to needs-triage for a maintainer. Triage it by hand, and read the runs for why intake did not:

  10. 1 remaining item

  11. melodic-automation-lanes commented on Nov 17, 2026

    @melodic-automation-lanes
    Contributor

    Not split at intake (owner Q93). This issue stays parked at status: ready: no split into whole changes was found, either by task-list line or by the targets the lines name. No person is asked to act and no hold label is added. The split settings are size-split in .github/actions/resolve-config/admission-rules.json.

    Automated by the lanes App, once per reason and body version.

  12. kyle-sexton commented on Nov 21, 2026

    @kyle-sexton
    MemberAuthor

    Split by hand into #17755, #17758, #17759, #17760, #17761 (native sub-issues, each Small or Medium, each eval-gated): intake sized this issue Large and its size split declined no-split. Most of the Large rating came from the "References outside the plugin" list, which was built by bare file name and pulls in about 300 unrelated files. Each child lists only the outside references a full-path git grep finds. Each child carries its rows, the rules and the host-session eval method, and closes with its own pull request. This issue's AC1 is met when every child closes by merge; the per-PR done-when lines live on the children.

  13. removed
    agent-readyFully specified and briefed; eligible for autonomous pickup from the frontier.
    on Nov 21, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    Fields

    Work status

    needs-triage

    Priority

    medium

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions