Skip to content

Cut the model-read ecosystem specifics from implementation #16762

Open

Description

@kyle-sexton

Parent

Refs #15703
Source: #15730 (blind classification under rules revision 5, #15730 comment 6950077480), cut item 10 of 47 in its filing order.

Depends on

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

Outcome

implementation ships no model-read tool, library or command choice outside ADR 0101's named exceptions: the 14 rows below are cut (R11 14).

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): 14 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
F0223 R11 K7 plugins/implementation/skills/implement/SKILL.md:137 implement SKILL.md:139 still lists eslint-disable, # noqa, # type: ignore, #pragma warning disable as model-read examples of a language-neutral no-suppression rule (quality, not safety).
F0224 R11 K7 plugins/implementation/skills/implement/SKILL.md:142; plugins/implementation/skills/implement/context/feature.md:13 SKILL.md:145 and feature.md:13 still give "a layered .NET/Clean-Architecture app" as a model-read example of dependency direction.
F0225 R11 K7 plugins/implementation/skills/implement/SKILL.md:159-161 The Design-signals table still gives # type: ignore[...] as a model-read example of weakening a type.
F0226 R11 K7 plugins/implementation/skills/implement/context/feature.md:37 feature.md:37 still describes C# MCP-server and .NET project-template scaffolding skills as model-read examples.
F0227 R11 K7 plugins/implementation/skills/implement/context/refactor.md:16 refactor.md:16 still names tsc and mypy as model-read examples of signature-only checks.
F0228 R11 K7 plugins/implementation/skills/implement/context/replace.md:25-27 replace.md:26 still names Jest property matchers and Verify scrubbers as model-read examples of scrubbing volatile fields.
F0229 R11 K5 plugins/implementation/skills/implement/context/replace.md:9,39,70-78 replace.md:9,39,70-73 point at Scientist, Jest, Verify and Istio pages, which are single-ecosystem tool documentation that R9 excludes, so they fall to model-read K5.
F0230 R11 K7 plugins/implementation/agents/phase-verifier.md:34-38 phase-verifier.md:34-36 still lists conftest.py, pytest ini, Jest/Vitest config and package.json test scripts as model-read examples in a test-tampering (quality) check, not a security-harm guard.
F0231 R11 K7 plugins/implementation/skills/implement-dispatch/SKILL.md:101 implement-dispatch SKILL.md item 13 still quotes the same model-read test-file example list verbatim.
F0232 R11 K7 plugins/implementation/skills/implement-dispatch/reference/holdout-tests.md:63-64 holdout-tests.md:61 still gives python -c, node -e as model-read examples of file access deny rules miss.
F0233 R11 K7 plugins/implementation/skills/implement-dispatch/reference/holdout-tests.md:84-95 holdout-tests.md:79-84 still gives pytest/vitest/jest invocations and config names as model-read examples of running a runner with holdout config.
F0234 R11 K5 plugins/implementation/skills/implement-dispatch/reference/holdout-tests.md:96-102 holdout-tests.md:90-93 points at pytest, Vitest and Jest CLI docs, single-ecosystem tool documentation that R9 excludes.
F0235 R11 K7 plugins/implementation/skills/implement-dispatch/SKILL.md:298 implement-dispatch SKILL.md:289 still gives "an assertEqual became an assertIn" as a model-read K7 example.
F0236 R11 K7 plugins/implementation/skills/implement-dispatch/SKILL.md:300 implement-dispatch SKILL.md:291 (moved from 300) still gives make -C and npm --prefix as model-read examples of directory flags.

Kept rows, not touched here (count per rule): none.

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 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.
  • References outside the plugin: these files on main name a cut file; update or remove each entry the cut makes stale: plugins/implementation/skills/implement/context/replace.md in scripts/docs-anchor-baseline.txt; plugins/implementation/skills/implement-dispatch/SKILL.md in scripts/docs-anchor-baseline.txt; plugins/implementation/skills/implement-dispatch/reference/holdout-tests.md in scripts/docs-anchor-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).
  • 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.

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): implementation ships no model-read tool, library or command choice outside the named exceptions; each listed R11 row is cut.
  • The pull request body states the row count (14) 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. 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).

  5. 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
  6. 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
  7. 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/44156896822

  8. 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:

  9. github-actions commented on Nov 17, 2026

    @github-actions
    Contributor

    Size: Medium
    Complexity: Medium
    Uncertainty: Known
    Flags: none
    Blocked by: none
    Blast radius: the implementation plugin's implement and implement-dispatch skills, the phase-verifier agent, and three entries in scripts/docs-anchor-baseline.txt; the model reads these skills and the agent at run time.
    Reversible: yes; revert the pull request.
    Open questions: none
    I read this as a bounded cut of 14 listed R11 rows of model-read ecosystem examples and tool pointers from the implementation plugin. Each row is deleted or replaced with generic text at a routing site, and the pull request body reports the row count, script-reader searches, eval cases and whole-or-part status. The route, the done-when and the dependencies are already stated, so nothing is missing. No duplicates or merged fixes found.

    Automated triage. A maintainer reviews it.

  10. 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.

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

    agent-readyFully specified and briefed; eligible for autonomous pickup from the frontier.

    Type

    Fields

    Work status

    ready

    Priority

    medium

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions