You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
docs-naming: read config from the config-cascade layers, not ~/.claude #18002
docs-naming still reads its user-global config from ~/.claude, including the retired docs-hygiene name. Move it to the config-cascade layers.
Current behavior
plugins/docs-naming/scripts/resolve-config.sh:
:11# (~/.claude/docs-naming.json), team (<home>/docs-naming.yaml), and a gitignored
:24-27 "RETIRED NAME. A user-global file absent under docs-naming but present under the retired docs-hygiene name (~/.claude/docs-hygiene.json) is read from there; resolve warns on stderr and paths reports the state legacy."
skills/setup/scripts/setup-apply.sh and setup-check.sh use the same paths. The team layer already reads <home>/docs-naming.yaml.
Desired behavior
Per the config-cascade layers, ~/.claude/docs-naming.json and ~/.claude/docs-hygiene.json move into userConfig (no user-global file layer). The resolver and setup read and write the contract layers; a file at either old path keeps being read during a dual-read window (see the acceptance criteria) and is reported with a warning. The canonical location list (docs/contracts/config-cascade/locations.md) lands in the docs/config-locations pull request.
Key interfaces
resolve-config.shresolve and paths output, including the legacy state
setup-apply.sh and setup-check.sh paths
Acceptance criteria
The resolver reads the config-cascade layers (the layers).
Dual-read window: while an old-path file (~/.claude/docs-naming.json or ~/.claude/docs-hygiene.json) exists, it is still read and its values still apply where no new-layer file supplies that layer; each run prints one warning naming the old file and the new location to move it to. No setting in an old file is silently dropped.
The plugin's setup (setup-check.sh, setup-apply.sh) names each old file it finds and offers to copy its keys into the new location after a preview and a yes, leaving the old file in place.
Removing the old-path read is a separate later change, made only once no old file is reported.
Tests cover: new layer only; old file only (values apply, warning printed); both present (new wins, warning names both).
Size: Medium
Complexity: Medium
Uncertainty: Known
Flags: none
Blocked by: none
Blast radius: the docs-naming plugin's resolve-config.sh, setup-apply.sh, setup-check.sh and their tests; anyone running docs-naming with a user-global config.
Reversible: yes; revert the change, since the old-path read stays during the dual-read window.
Open questions: none
I understood this as moving docs-naming's user-global config from ~/.claude/docs-naming.json and the retired docs-hygiene.json to the config-cascade layers. The old paths stay readable with a warning during a dual-read window, setup offers to copy keys over, and tests cover the three cases. The acceptance criteria are exact. Similar issues for other plugins are separate, as the issue says, so none of them is a duplicate, and I found no merged pull request that already does this.
Not admitted to the issue lane: its body was edited after it was deferred, so the admission it was deferred from judged an older body. Admission takes the issue once each of those is met, when intake triages it again (.github/actions/resolve-config/admission-rules.json).
This records a re-fire of an earlier stop, not a new one. The start gate stopped run 44952341755, the work-resolve-issue run that agent-ready started, before any model turn, because the GITHUB_TOKEN budget was under its start-gate share, or the window was projected to have under its start-project share of the header's limit left by the run's expected end (github-token start-gate, start-project and start-run-minutes in rate-limits.json; the run's start-gate notice names which); that run ended at 2026-11-21T21:37:12Z. work-unblock has now removed and added agent-ready back to start a fresh run. A budget stop is the lanes' load, not a fault in this issue: it never counts toward limits.limit-stops-per-issue or the kill-switch wait.
Activity
github-actions commentedon Nov 21, 2026
Size: Medium
Complexity: Medium
Uncertainty: Known
Flags: none
Blocked by: none
Blast radius: the docs-naming plugin's resolve-config.sh, setup-apply.sh, setup-check.sh and their tests; anyone running docs-naming with a user-global config.
Reversible: yes; revert the change, since the old-path read stays during the dual-read window.
Open questions: none
I understood this as moving docs-naming's user-global config from
~/.claude/docs-naming.jsonand the retireddocs-hygiene.jsonto the config-cascade layers. The old paths stay readable with a warning during a dual-read window, setup offers to copy keys over, and tests cover the three cases. The acceptance criteria are exact. Similar issues for other plugins are separate, as the issue says, so none of them is a duplicate, and I found no merged pull request that already does this.Automated triage. A maintainer reviews it.
melodic-automation-lanes commentedon Nov 21, 2026
{"v":1,"actor_id":337595936,"title_sha256":"b85661d311a39f26a19106a0e72b4495724e9d66242d4aa4cecc401f77064d00","body_sha256":"87ae80613adde474d72b4ff0282900e6ff7a17d79ebb545a39c7902fe39f9f77"}
melodic-automation-lanes commentedon Nov 21, 2026
Not admitted to the issue lane: its body was edited after it was deferred, so the admission it was deferred from judged an older body. Admission takes the issue once each of those is met, when intake triages it again (
.github/actions/resolve-config/admission-rules.json).melodic-automation-lanes commentedon Nov 21, 2026
{"v":1,"actor_id":337595936,"title_sha256":"b85661d311a39f26a19106a0e72b4495724e9d66242d4aa4cecc401f77064d00","body_sha256":"794dfe5fe4f105eb3259ef3bf9ba782611198fdb25997c4ddf22bca10f1441e5"}
melodic-automation-lanes commentedon Nov 21, 2026
This records a re-fire of an earlier stop, not a new one. The start gate stopped run 44952341755, the work-resolve-issue run that
agent-readystarted, before any model turn, because the GITHUB_TOKEN budget was under its start-gate share, or the window was projected to have under its start-project share of the header's limit left by the run's expected end (github-token start-gate, start-project and start-run-minutes in rate-limits.json; the run's start-gate notice names which); that run ended at 2026-11-21T21:37:12Z. work-unblock has now removed and addedagent-readyback to start a fresh run. A budget stop is the lanes' load, not a fault in this issue: it never counts toward limits.limit-stops-per-issue or thekill-switchwait.github-actions commentedon Nov 21, 2026