@jphutchins/code-review 0.1.0-alpha.66 → 0.1.0-alpha.68

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@jphutchins/code-review",
-  "version": "0.1.0-alpha.66",
+  "version": "0.1.0-alpha.68",
   "description": "Deterministic commenter for agentic PR review — gather, render, inline, post, adapt, extract, cost, validate, print-schema",
   "type": "module",
   "license": "MIT",

package/schema/VERSIONING.md

@@ -75,7 +75,7 @@ ignored); a version outside the supported set degrades to a §5.5 sticky notice.
 | `v0.6.0` | superseded | Adds optional `systemic_problems` — an array of cross-cutting observations that tie findings together and are hard to express with a line range, each item with required `title`/`description`/`severity`/`reasoning`/`confidence` and optional `code`/`code_url`/`finding_codes`/`paths` (no line anchors). Refocuses `summary` on justifying the overall verdict rather than restating findings. Backwards-compatible (a `0.5` document is a valid `0.6` document); the CLI keeps resolving `0.4`/`0.5` via identity upcasts. |
 | `v0.9.0` | superseded (frozen at `schema/v0.9/`) | Adds a REQUIRED `likelihood` (0..1) to every finding and systemic problem (issue #163): the probability the triggering input/state actually occurs — distinct from `confidence` (whether the defect is real) — folded into the convergence score as a second multiplier. **Breaking**: a pre-0.9 document lacks `likelihood` and no longer validates, so a stale sticky degrades to a sentinel-only seed (one cold re-review) rather than seeding. The draft axis skips `0.7`/`0.8`: those are the surface-signal axis's versions (below), so the draft jumps past them to keep the two version spaces distinct — a draft must never be mistaken for a legacy surfaced blob (issue #156). |
 | `v0.10.0` | superseded | Renames the mechanism identifier `code` → `id` and makes it **required** on every finding (issue #246), so every finding is always trackable across rounds and discussions. Systemic problems rename `code`/`finding_codes` → `id`/`finding_ids`; convergence rounds rename `codes` → `ids`; `scope_metastasis.recurring` renames `code` → `id`. `change_size.code` (the role bucket) and `code_url` keep their names. **Breaking**: a pre-0.10 document lacks `id` — the registry upcasts every pre-0.10 minor through one tolerant legacy codec (`code` → `id`, or a synthesized content-derived id — sha256 of path+title — when a finding carried none), so a legacy sticky still seeds a re-review with its ids intact, and the pre-0.10 shape stays frozen at `schema/v0.9/` (above). |
-| `v0.11.0` | **current** (draft) | Adds an optional `rebuttal` to every finding and systemic problem: on a re-raise, the reviewer's answer to the response recorded against that id (the review dialogue, #247). A re-raise that carries a non-blank `rebuttal` is never a verbatim re-raise. A rebuttal answers one round's response, so the re-review seed strips rebuttals from the prior review it carries forward, and a re-raise must supply its own. The legacy codec drops a `rebuttal` on a pre-0.10 document, which the frozen `v0.9` file rejects: the one deliberate codec-only tolerance, since only a hybrid document ever carries one. The systemic `id` description now states that a re-raise of the same class reuses the id. Backwards-compatible: a `0.10` document is a valid `0.11` document, so both minors resolve through the same live file and codec, the way `0.4`–`0.9` share the frozen legacy file. Draft until the next release, so the dialogue's later phases may still add to it. |
+| `v0.11.0` | **current** (draft) | Adds an optional `rebuttal` to every finding and systemic problem: on a re-raise, the reviewer's answer to the response recorded against that id (the review dialogue, #247). A re-raise that carries a non-blank `rebuttal` is never a verbatim re-raise. A rebuttal answers one round's response, so the re-review seed strips rebuttals from the prior review it carries forward, and a re-raise must supply its own. The legacy codec drops a `rebuttal` on a pre-0.10 document, which the frozen `v0.9` file rejects: the one deliberate codec-only tolerance, since only a hybrid document ever carries one. The systemic `id` description now states that a re-raise of the same class reuses the id. Adds an optional top-level `unfinished`: the checks a review planned but did not complete, rendered beside the convergence badge. Backwards-compatible: a `0.10` document is a valid `0.11` document, so both minors resolve through the same live file and codec, the way `0.4`–`0.9` share the frozen legacy file. Draft until the next release, so the dialogue's later phases may still add to it. |
 
 ### Surface channel (stop signal)
 

package/schema/findings.schema.json

@@ -176,6 +176,11 @@
         }
       }
     },
+    "unfinished": {
+      "type": "array",
+      "description": "Checks this review planned but did not complete, one entry each.",
+      "items": { "type": "string", "pattern": "\\S" }
+    },
     "change_size": {
       "type": "object",
       "additionalProperties": false,

package/templates/comment.eta

@@ -47,6 +47,14 @@
 <% } -%>
 
 <%~ it.findings.summary %>
+<% if (it.unfinished.length > 0) { -%>
+
+> [!WARNING]
+> **Unfinished checks** — this round did not complete every check it planned, so its convergence covers less than a full review:
+<% it.unfinished.forEach(function(check) { -%>
+> - <%~ check %>
+<% }) -%>
+<% } -%>
 <% const c = it.severityCounts %><% const findingsTotal = c.critical + c.major + c.minor + c.nit -%>
 <% if (it.strays.length > 0 || (!it.incomplete && (findingsTotal > 0 || it.systemic.length > 0 || it.contested.length > 0))) { -%>
 
@@ -160,9 +168,9 @@ _The reviewer rebutted a maintainer's answer, and a maintainer answered again. E
 **Findings:** <%~ parts.join(' · ') %>
 <% } else if (!it.incomplete && (it.systemic.length > 0 || it.contested.length > 0)) { %>
 <% const held = [] %><% if (it.contested.length > 0) { held.push(it.contested.length + ' contested item' + (it.contested.length === 1 ? '' : 's')) } %><% if (it.systemic.length > 0) { held.push(it.systemic.length + ' systemic problem' + (it.systemic.length === 1 ? '' : 's')) } %>
-_No findings beyond the <%= held.join(' and ') %> above._
+_No findings beyond the <%= held.join(' and ') %> above<%= it.unfinishedCaveat %>._
 <% } else if (!it.incomplete && !it.answeredReRaiseNote) { %>
-<% if (it.unverifiedNoLogs) { %>_No findings — but no failing-job logs were staged, so this pass could not read the CI failure. "No findings" is not evidence of none._<% } else { %>_No findings — clean review._<% } %>
+<% if (it.unverifiedNoLogs) { %>_No findings — but no failing-job logs were staged, so this pass could not read the CI failure. "No findings" is not evidence of none._<% } else if (it.unfinishedCaveat) { %>_No findings in the checks this round completed<%= it.unfinishedCaveat %>._<% } else { %>_No findings — clean review._<% } %>
 <% } %>
 
 <% if (it.inlineDisposition && it.inlineDisposition.kind === 'posted') { %>