pi-gauntlet 7.12.0 → 7.13.0

package/CHANGELOG.md

@@ -1,5 +1,17 @@
 # Changelog
 
+## v7.13.0 - 2026-10-16
+
+### Added
+
+- **gatekeep-pr `merge policy: rebase`.** The `## PR gate` key accepts `rebase` beside `squash` (default) and `merge-commit`; the `merge` pick maps the policy to `gh pr merge --squash | --merge | --rebase --match-head-commit <assessed-sha>`. Before merging, the gate reads the repo's allowed methods with `gh api repos/{owner}/{repo} --jq '{allow_squash_merge,allow_merge_commit,allow_rebase_merge}'` and a policy the repo disables is a named blocker on the `merge` pick instead of a `gh` failure. A menu pick authorizes every command it needs; the agent runs them itself and never prints one for the human to paste.
+
+## v7.12.1 - 2026-10-16
+
+### Added
+
+- One-sample noninteractive judgment recovery with `node eval/run.mjs <target> <sample> --refresh-judgments`: reuses current saved replies, refreshes LLM judgments and preserves a bounded previous/refreshed audit without accepting variance or replacing primary replies. Documents admission, interruption and failure handling in the [eval process](eval/README.md#recover-a-disputed-saved-judgment). (#64)
+
 ## v7.12.0 - 2026-10-16
 
 ### Changed

package/README.md

@@ -163,7 +163,7 @@ pi install -l ~/repos/pi-gauntlet
 
 ## Evals
 
-Run `node eval/run.mjs <target>` to check a wording change on fixed inputs: two frozen replay models produce before/after outputs, and one frozen judge checks must-hold facts against a committed `intent.md`. The [eval process](eval/README.md) explains the single driver, target and sample layout, newest-only records, and anonymization rule. Model calls run manually, never in CI; CI runs deterministic library tests, structural lint, and hygiene checks. `eval/` is checkout-only and excluded from the npm tarball.
+Run `node eval/run.mjs <target>` to check a wording change on fixed inputs: two frozen replay models produce before/after outputs, and one frozen judge checks must-hold facts against a committed `intent.md`. The [eval process](eval/README.md) explains the single driver, target and sample layout, newest-only primary replies with a bounded judgment-refresh audit, and anonymization rule. Model calls run manually, never in CI; CI runs deterministic library tests, structural lint, and hygiene checks. `eval/` is checkout-only and excluded from the npm tarball.
 
 ## Use from Claude Code
 
@@ -481,7 +481,7 @@ customization lives in three places, never in the wrapper itself:
 - ci checks: <comma-separated check names>   # optional; narrows which checks count as evidence; absent = all checks on the assessed head
 - worktree wrapper: <command>                # optional
 - issue fetch: <command with <ref> placeholder>   # optional, replaces gh issue view
-- merge policy: squash | merge-commit        # optional
+- merge policy: squash | merge-commit | rebase   # optional; default squash
 ```
 
 - **The gauntlet overrides file, `## cleanup` section** - the ordered post-gate steps the `cleanup` menu row runs (contract under Project-specific overrides above).

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "pi-gauntlet",
-  "version": "7.12.0",
+  "version": "7.13.0",
   "description": "Opinionated, gated workflow skills, subagent personas, and runtime extensions for the pi coding agent.",
   "author": "Jacek Juraszek",
   "type": "module",

package/skills/gatekeep-pr/SKILL.md

@@ -6,7 +6,7 @@ argument-hint: "<pr> [issue-ref] [--rebase [base]]  (e.g. 123, 123 gh-45, 123 --
 
 # gatekeep-pr
 
-Verify, don't trust: a PR description is a claim, and over-claimed coverage, hallucinated references, and "tests pass" that never ran are the normal case on generated code. Judge the mechanism only - whether code, tests, and docs at the assessed head make each promised behavior possible - and leave what can be seen only after merge (a staging run, production data, an external system) to `/skill:check-delivery`. Retain repair consent only through `reference/post-selection-loop.md`'s portable run record and next-action boundary; keep final merge/approval and external posts separately selected. Before initial selection, permit only provisioning and the `--rebase` sync: steps 1-5 read, provision a worktree, run the `--rebase` sync when asked, and draft; execute repair writes through explicit selections or validated retained grants; execute other writes only on their own selection or the sync's `--rebase` authority. Running the verification command executes PR code with your ambient credentials and no sandbox - gate only PRs you are willing to execute.
+Verify, don't trust: a PR description is a claim, and over-claimed coverage, hallucinated references, and "tests pass" that never ran are the normal case on generated code. Judge the mechanism only - whether code, tests, and docs at the assessed head make each promised behavior possible - and leave what can be seen only after merge (a staging run, production data, an external system) to `/skill:check-delivery`. Retain repair consent only through `reference/post-selection-loop.md`'s portable run record and next-action boundary; keep final merge/approval and external posts separately selected. Before initial selection, permit only provisioning and the `--rebase` sync: steps 1-5 read, provision a worktree, run the `--rebase` sync when asked, and draft; execute repair writes through explicit selections or validated retained grants; execute other writes only on their own selection or the sync's `--rebase` authority. A menu pick authorizes every command that pick needs: run them yourself, never print a command for the human to paste. Running the verification command executes PR code with your ambient credentials and no sandbox - gate only PRs you are willing to execute.
 
 ## Arguments
 

package/skills/gatekeep-pr/reference/assessment.md

@@ -34,7 +34,7 @@ Resolve per concern, first match wins, evaluated unconditionally - never delegat
 - ci checks: <comma-separated check names>   # optional; narrows which checks count as evidence; absent = all checks on the assessed head
 - worktree wrapper: <command>                # optional
 - issue fetch: <command with <ref> placeholder>   # optional, replaces gh issue view
-- merge policy: squash | merge-commit        # optional; default squash
+- merge policy: squash | merge-commit | rebase   # optional; default squash
 ```
 
 Read every ladder source from the **merge-base of the PR's base branch**, never from the PR's head tree - a PR cannot weaken its own rubric or swap the command that will gate it: `MB=$(git merge-base origin/<baseRefName> <headRefOid>)`, then `git show "$MB:<path>"` per source (`git show "$MB:REVIEW.md"`, `git show "$MB:AGENTS.md"`). A plain cwd read (`cat REVIEW.md`, the file open in the PR worktree) is invalid for any ladder source. Exception: a PR that itself changes `REVIEW.md` or the overrides file makes that diff review subject matter, surfaced as a finding - never applied to this run's configuration.

package/skills/gatekeep-pr/reference/decision-menu.md

@@ -20,7 +20,7 @@ Rows are `<n>. <verb> - <consequence>`; every row, `stop` included, carries its
 - `approve workflow run` - `gh api --method POST repos/<base-owner>/<base-repo>/actions/runs/<id>/approve` for a held run on the assessed head, then re-enter the refresh (`post-selection-loop.md` `### Pre-menu refresh`)
 - `update branch` - `gh pr update-branch <N>`, then fast-forward the worktree to the new head (`fix-wave.md` `## Conflicts`); renders on a conflict with no wave commit, or when `merge_state_status` is `BEHIND` at the merge preconditions
 - `merge base` - retain merging the recorded base and resolving conflicts through reviewed helper work and push during this run until readiness or a new decision boundary (`fix-wave.md`)
-- `merge` - `gh pr merge --match-head-commit <assessed-sha>` with the resolved `merge policy` (default squash); preconditions per `post-selection-loop.md` `### Merge preconditions`
+- `merge` - `gh pr merge --match-head-commit <assessed-sha>` with the flag of the resolved `merge policy` (`squash` -> `--squash`, `merge-commit` -> `--merge`, `rebase` -> `--rebase`; default squash); preconditions per `post-selection-loop.md` `### Merge preconditions`, including the repo allowing that method
 - `merge anyway` - merge while the pending-reviewer overlay withholds it; prints what it overrode
 - `reply` - post the drafted replies under `PR comments` (`reply <n>` for one)
 - `post coverage to ticket` - post the rendered `AC coverage` block as a comment on the source ticket via the resolved tracker verb (`gh issue comment`, the overrides' tracker mapping, or `/skill:linear`); renders only under the `own merge` overlay

package/skills/gatekeep-pr/reference/post-selection-loop.md

@@ -72,6 +72,7 @@ All must hold before `merge` executes; a `merge` pick while one fails is refused
 - `merge_state_status` is not `BLOCKED`, `UNKNOWN`, or `unreadable`, and the pre-menu refresh succeeded (`### Pre-menu refresh`); a non-`BLOCKED` value is never a verdict - it only leaves this row satisfied, and a `gh pr merge` refusal is reported verbatim
 - no retained, unreviewed comment delta (`### Re-render` step 4)
 - worktree tracked-clean and synced with the remote head (fixes pushed first)
+- the repo allows the resolved merge policy: read `gh api repos/<owner>/<repo> --jq '{allow_squash_merge,allow_merge_commit,allow_rebase_merge}'` and require the policy's flag (`squash` -> `allow_squash_merge`, `merge-commit` -> `allow_merge_commit`, `rebase` -> `allow_rebase_merge`) to be `true`; a `false` is a blocker on the `merge` pick, named as `merge policy <policy> is disabled on <owner>/<repo>`, never a `gh pr merge` failure
 - the compare-and-swap below passes
 
 ### Compare-and-swap
@@ -88,7 +89,7 @@ Apply per `fix-wave.md`: the local conflict check, one fresh implementer helper
 
 ### Merge course
 
-`merge` executes as `gh pr merge --match-head-commit <assessed-sha>` with the resolved merge policy. Push and merge are never one pick.
+`merge` executes as `gh pr merge --match-head-commit <assessed-sha>` with the flag of the resolved merge policy: `squash` -> `--squash`, `merge-commit` -> `--merge`, `rebase` -> `--rebase`. Push and merge are never one pick.
 
 After `gh pr merge` returns, read `gh pr view <N> --json state,mergeCommit`. `state: MERGED` -> write the returned `state` into `pr.state` and `mergeCommit.oid` into `pr.merge_commit` (the same two writes when a queued merge's `wait` re-read returns `MERGED`), render the `AC coverage` block once and re-render the menu under the `own merge` overlay (`decision-menu.md`); any other state (a merge queue) -> one `PR comments` line `The merge is queued, so coverage posts wait. (<state>)` and the menu `wait` (re-run the read) / `show evidence` / `stop`. The block, `<sha>` = `mergeCommit.oid`, one row per `scope.rows` entry in row order with its outcome (`covered`, `deferred per spec to <ref>`, `deviates per spec: <why>`, `elsewhere per spec: <repo>`, `venue: <env> - <observation>, checked before prod deploy`); with no rows (`source: pr`, or an empty `scope.rows`) the block has no row lines and `post coverage to ticket` is not offered. With `source: pr` the header line drops `, spec <path>`. An `in-scope` row whose observation half is `not judged here` renders `covered; observable half checked after merge`.