gaurav0107 · gaurav0107 · Jul 21, 2026 · Jul 20, 2026 · Jul 20, 2026 · Jul 20, 2026
diff --git a/agents/repo-finder.md b/agents/repo-finder.md
@@ -575,7 +575,7 @@ fi
   2. A `good first issue` or `help wanted` label (explicit triage-accepted signal).
   3. A `kind:*` / `type:*` / `priority:*` / `severity:*` / `area:*` label (maintainer-applied taxonomy — NOT just `needs-triage` / `triage` / `status:need-triage` alone).
 
-  Why this gate exists: a `needs-triage`-only issue with zero maintainer comments is an untriaged feature request or unconfirmed bug report. Shipping a PR against it at Apache-/Airflow-class repos triggers a lazy-consensus or devlist redirect that structurally caps merge probability below 95% regardless of code quality. The prior `has_maintainer_approval` field was additive (+3 bonus), so issues without maintainer approval could still reach rank 1 via the other bonuses. This gate converts it to a prerequisite.
+  Why this gate exists: a `needs-triage`-only issue with zero maintainer comments is an untriaged feature request or unconfirmed bug report. Shipping a PR against it at Apache-/Airflow-class repos triggers a lazy-consensus or devlist redirect that structurally caps merge probability below 95% regardless of code quality. The maintainer-signal bonus is additive, so an issue no maintainer engaged with could otherwise reach rank 1 on the other bonuses alone. This gate converts engagement to a prerequisite.
 
   A bot is not a maintainer — exclude bot accounts from the comment test.
 
@@ -726,7 +726,7 @@ jq -r --rawfile m "$MAINTAINERS" \
     elif ((now - ($engaged | max)) / 86400 | floor) > $window then
       "SKIP \(.number): no maintainer engagement in \(((now - ($engaged | max)) / 86400 | floor))d (window \($window)d) — abandoned"
     else
-      "KEEP \(.number) engaged_days=\(((now - ($engaged | max)) / 86400 | floor)) maintainer_commented=\(.maintainer_commented) approval=\(.maintainer_comment_assoc)"
+      "KEEP \(.number) engaged_days=\(((now - ($engaged | max)) / 86400 | floor)) maintainer_commented=\(.maintainer_commented) signal=\(.maintainer_signal)"
     end
 ' "$SCRATCH/stage_a.jsonl"
 ```
@@ -751,14 +751,30 @@ visibly drop one. Apply the prose here, then say in `notes` that it did.
 - **Is a bug** (not a feature request): +3
 - **Has reproduction steps**: +2
 - **Labeled good-first-issue or help-wanted**: +2
-- **Maintainer commented with approval signal** ("PRs welcome", "happy to review"): +3. Requires `maintainer_comment_assoc` — a union-only commenter does not earn this bonus.
+- **Maintainer signal**, graded from `maintainer_signal`: `invites_pr` **+5** · `confirms` **+3** · `neutral` **+1** · `none` **0**. Association is the prerequisite, never the grade — a union-only commenter earns nothing here.
 - **Maintainer engaged within the last 28 days**: +2
 - **Issue age**, graded: 2–30d **+1** · 30–90d **0** · 90–365d **−1** · >365d **−2**. Scored, never hard-skipped — old is riskier, not worthless.
 - **Scope is small** (likely < 100 lines, single-file fix): +2
 
-Set `best_issue.has_maintainer_approval` from `maintainer_comment_assoc` — true only when GitHub reported a commenter on this issue as `OWNER`/`MEMBER`/`COLLABORATOR`. Never set it from `maintainer_commented`: that flag includes the wide 3c maintainer union, which is correct for passing the gate but too loose to call approval.
-
-Label-only triage and union-only comments still pass the gate; they just record `has_maintainer_approval=false`. The field answers one question — "did an account GitHub vouches for as a maintainer speak on this issue?" — and the +3 approval bonus is the only thing that reads it.
+Set `best_issue.maintainer_signal` from the `maintainer_signal` the triage filter
+emitted. Do not recompute it from `maintainer_comment_assoc`: association answers
+"did an account GitHub vouches for speak here", which is a prerequisite for being
+graded and not a grade. Never derive it from `maintainer_commented` either — that
+flag includes the wide 3c union, correct for passing the gate and too loose to
+read as endorsement.
+
+The filter grades on comment text, and only on text that is text: it strips links
+and bare mentions first, so a maintainer comment that is only a pointer somewhere
+else grades `none`. A pointer states no position on this issue.
+
+Read the graded comment yourself before accepting the tier. The filter matches
+prose with regexes and will misgrade: a refutation phrased in the project's own
+domain terms disputes the premise while matching none of the decline patterns.
+**If the maintainer disputes the premise, redirects it, or calls it
+working-as-intended, skip the issue outright** rather than scoring it down —
+zeroing the bonus still leaves bug + repro + small scope at 7, and labels plus
+recency clear the 8 threshold on their own. Record any tier you overrode in
+`notes`, so the regex gaps stay visible.
 
 If no issue scores 8+ after applying the hard filter, mark the repo as "no clear opportunity right now" and exclude from the final list.
 
@@ -854,7 +870,7 @@ needs that number to know the list is partial rather than the field being thin.
         "title": "issue title",
         "type": "bug",
         "labels": ["bug", "good first issue"],
-        "has_maintainer_approval": true,
+        "maintainer_signal": "invites_pr",
         "issue_score": 12,
         "url": "https://github.com/OWNER/REPO/issues/1234"
       },
@@ -880,7 +896,7 @@ Found N repos worth contributing to (searched M candidates).
 
 ## Top Pick
 **owner/repo#1234** — [issue title]
-- Maintainer approved: yes
+- Maintainer signal: invites a PR
 - Type: bug
 - Estimated scope: small (single file)
 - Median PR merge time: 2 days

diff --git a/schemas/repo_shortlist.schema.json b/schemas/repo_shortlist.schema.json
@@ -80,7 +80,10 @@
               "title":                   {"type": "string"},
               "type":                    {"type": "string"},
               "labels":                  {"type": "array", "items": {"type": "string"}},
-              "has_maintainer_approval": {"type": "boolean"},
+              "maintainer_signal": {
+                "type": "string",
+                "enum": ["invites_pr", "confirms", "neutral", "none"]
+              },
               "issue_score":             {"type": "number"},
               "url":                     {"type": "string"}
             }

diff --git a/scripts/orchestrator/triage_filter.sh b/scripts/orchestrator/triage_filter.sh
@@ -75,7 +75,31 @@ jq -c --rawfile m "$MAINTAINERS" --argjson now "$NOW" '
   + "|i.m working on|working on (a fix|this)"
   + "|(already |should have )?fixed (it |this )?in [a-z0-9.@/#-]"
   + "|(fix|patch) (has )?landed|landed in|resolved (in|by) (#|[a-z0-9])"
-  + "|(fix|pr) (is )?incoming")                                             as $claimed
+  + "|(fix|pr) (is )?incoming"
+  # A claim is often worded as an intention to act rather than a patch in hand.
+  # "this is valid, we will look into it" confirms the defect and takes it in one
+  # sentence — which is why the claim test runs before the signal tiering below.
+  # "on our side" only claims the work when it is attached to an intent to act:
+  # "the regression is on our side" owns the defect and invites a fix.
+  + "|we.ll look into|will look into|looking into (this|it) on our"
+  + "|noted (it )?(on our side|internally)|taking (this|it) on")             as $claimed
+
+  # An outsider saying "I would like to work on this" is not an assignee, so the
+  # assignee test cannot see the claim. Windowed, unlike the maintainer test: a
+  # drive-by claim nobody acted on must not fence the issue off forever.
+  | ("i.d like to work on|i would like to work on|can i (take|work on|pick up)"
+  + "|i.ll (take|work on|pick) (this|it)|assign (this|it) to me"
+  + "|i.m (going to |gonna )?work(ing)? on (this|it)"
+  + "|(may|could) i (take|work on) (this|it)")                      as $outsider_claim
+  | 1209600                                                         as $claim_ttl
+
+  # Association makes a comment gradeable; it does not make it approval. The text
+  # sets the grade.
+  | ("prs? (are )?welcome|welcome a (pr|patch)|happy to (review|take a pr|accept)"
+  + "|feel free to (submit|open|send)|would (accept|welcome) a (pr|patch)"
+  + "|go ahead and (open|submit)|contributions? welcome")                 as $invites
+  | ("reproduc|confirmed|i see the same|can confirm|this is (valid|a bug|indeed)"
+  + "|good catch|you.re right|nice find|makes sense to me")               as $confirms
 
   # Announcements and containers wear defect labels. A pinned "the project moved"
   # notice earns a maintainer taxonomy label and every engagement signal the gate
@@ -104,6 +128,35 @@ jq -c --rawfile m "$MAINTAINERS" --argjson now "$NOW" '
   | ($i.labels | map(.name | ascii_downcase))                              as $L
   | ($mc | map(.body // "") | join("\n") | ascii_downcase)                 as $mbody
 
+  # Grade only what is left after stripping links and bare @mentions: a comment
+  # that is only a pointer somewhere else states no position on this issue, and no
+  # sentiment tier can read one.
+  | ($mca | map((.body // "")
+                | ascii_downcase
+                | gsub("https?://\\S+"; " ")
+                | gsub("@[a-z0-9_-]+"; " ")
+                | gsub("\\s+"; " ") | sub("^ +"; "") | sub(" +$"; ""))
+          | map(select(length > 0)))                                    as $stripped
+  | ($stripped | join("\n"))                                            as $gbody
+  # Match before measuring. The floor separates a remark from a pointer, so it
+  # decides neutral-or-none only — applied first it would grade "PRs welcome!"
+  # (12 chars, the strongest signal there is) as no signal at all.
+  | (if   ($stripped | length) == 0          then "none"
+     elif ($gbody | test($invites))          then "invites_pr"
+     elif ($gbody | test($confirms))         then "confirms"
+     elif ($stripped | any(length >= 15))    then "neutral"
+     else "none" end)                                                   as $signal
+
+  # Non-maintainer claims, still inside the window.
+  | ($human | map(select((.authorAssociation | IN("OWNER","MEMBER","COLLABORATOR")) | not))
+           # `// ""` is not enough: an undated comment must not reach
+           # fromdateiso8601, which throws on null and aborts the whole batch.
+           # Undated reads as old, so the claim expires rather than fencing off.
+           | map(select(((.body // "") | ascii_downcase | test($outsider_claim))
+                        and (((.createdAt // "") | length) > 0)
+                        and ((.createdAt | fromdateiso8601) > ($now - $claim_ttl))))
+           | length)                                                    as $outsider_live
+
   # (24h rule) an issue younger than a day has not been triaged; racing to it
   # produces noise PRs.
   | select(($i.createdAt | fromdateiso8601) <= ($now - 86400))
@@ -139,6 +192,9 @@ jq -c --rawfile m "$MAINTAINERS" --argjson now "$NOW" '
     elif ($mbody | test($claimed)) then
       {verdict: "SKIP", number: $i.number,
        reason: "maintainer is already fixing it"}
+    elif $outsider_live > 0 then
+      {verdict: "SKIP", number: $i.number,
+       reason: "claimed — another contributor said they are taking it"}
     else
       # Carry the issue payload, not just a verdict on a number. The Step-4 rubric
       # scores title, labels, body and createdAt, and the agent contract forbids
@@ -150,6 +206,7 @@ jq -c --rawfile m "$MAINTAINERS" --argjson now "$NOW" '
        createdAt: $i.createdAt,
        maintainer_commented: ($mc | length > 0),
        maintainer_comment_assoc: ($mca | length > 0),
+       maintainer_signal: $signal,
        last_maintainer_comment: ($mc | map(.createdAt) | max)}
     end
 ' "$ISSUES"
diff --git a/tests/scripts/test_triage_filter.sh b/tests/scripts/test_triage_filter.sh
@@ -13,6 +13,8 @@ trap 'rm -rf "$tmpdir"' EXIT
 NOW=1783814400
 OLD="2026-07-01T00:00:00Z"   # 11 days old — clears the 24h rule
 FRESH="2026-07-11T18:00:00Z" # 6 hours old  — must be dropped
+RECENT="2026-07-10T00:00:00Z" # 2 days old  — a claim this fresh is still live
+STALE="2026-05-20T00:00:00Z"  # 53 days old — a claim this old is abandoned
 
 printf 'unionlead\n' > "$tmpdir/maintainers.txt"
 
@@ -106,7 +108,59 @@ cat > "$tmpdir/issues.json" <<EOF
   {"number": 913, "title": "[v4] Loader raises on a valid checkpoint",
    "createdAt": "$OLD", "assignees": [], "labels": [{"name": "bug"}],
    "comments": [{"authorAssociation": "MEMBER", "author": {"login": "maint-b"},
-     "body": "We can't fix bugs on v4 — please upgrade to v5.", "createdAt": "$OLD"}]}
+     "body": "We can't fix bugs on v4 — please upgrade to v5.", "createdAt": "$OLD"}]},
+
+  {"number": 914, "title": "Indexer produces incomplete examples for one doc format",
+   "createdAt": "$OLD", "assignees": [], "labels": [{"name": "bug"}],
+   "comments": [{"authorAssociation": "COLLABORATOR", "author": {"login": "maint-d"},
+     "body": "Thanks for the report, this is valid. We have noted it on our side and will look into supporting that format.",
+     "createdAt": "$OLD"}]},
+
+  {"number": 915, "title": "Batched decode corrupts every row after the first",
+   "createdAt": "$OLD", "assignees": [], "labels": [{"name": "kind:bug"}],
+   "comments": [{"authorAssociation": "NONE", "author": {"login": "eager-newcomer"},
+     "body": "Hi! I'd like to work on this issue if it's still available.",
+     "createdAt": "$RECENT"}]},
+
+  {"number": 916, "title": "Same claim, long abandoned",
+   "createdAt": "$OLD", "assignees": [], "labels": [{"name": "kind:bug"}],
+   "comments": [{"authorAssociation": "NONE", "author": {"login": "eager-newcomer"},
+     "body": "Hi! I'd like to work on this issue if it's still available.",
+     "createdAt": "$STALE"}]},
+
+  {"number": 917, "title": "Usage totals are summed across tool-loop calls",
+   "createdAt": "$OLD", "assignees": [], "labels": [{"name": "bug"}],
+   "comments": [{"authorAssociation": "COLLABORATOR", "author": {"login": "maint-d"},
+     "body": "https://example.invalid/acme/repo/pull/1", "createdAt": "$OLD"}]},
+
+  {"number": 918, "title": "Quantized matmul ignores the leading batch dimension",
+   "createdAt": "$OLD", "assignees": [], "labels": [{"name": "bug"}],
+   "comments": [{"authorAssociation": "MEMBER", "author": {"login": "maint-b"},
+     "body": "cc @maint-c for the quantization path", "createdAt": "$OLD"}]},
+
+  {"number": 919, "title": "Join is unsanitised on the secondary path",
+   "createdAt": "$OLD", "assignees": [], "labels": [{"name": "bug"}],
+   "comments": [{"authorAssociation": "MEMBER", "author": {"login": "maint-b"},
+     "body": "Reproduced on main — the second join has the same defect.",
+     "createdAt": "$OLD"}]},
+
+  {"number": 920, "title": "Terse blessing is still a blessing",
+   "createdAt": "$OLD", "assignees": [], "labels": [{"name": "bug"}],
+   "comments": [{"authorAssociation": "MEMBER", "author": {"login": "maint-b"},
+     "body": "PRs welcome!", "createdAt": "$OLD"}]},
+
+  {"number": 921, "title": "Owning the defect is not claiming the fix",
+   "createdAt": "$OLD", "assignees": [], "labels": [{"name": "bug"}],
+   "comments": [{"authorAssociation": "MEMBER", "author": {"login": "maint-b"},
+     "body": "Thanks — reproduced. The regression is on our side, not in your config.",
+     "createdAt": "$OLD"}]},
+
+  {"number": 922, "title": "Comment with no timestamp must not abort the batch",
+   "createdAt": "$OLD", "assignees": [], "labels": [{"name": "bug"}],
+   "comments": [{"authorAssociation": "NONE", "author": {"login": "eager-newcomer"},
+      "body": "I would like to work on this."},
+     {"authorAssociation": "MEMBER", "author": {"login": "maint-b"},
+      "body": "Reproduced, the parser drops the last byte.", "createdAt": "$OLD"}]}
 ]
 EOF
 
@@ -252,4 +306,45 @@ esac
 [ "$(field 21795 title)" = "null" ] \
   || { echo "FAIL #21795 is a SKIP and must not carry a title"; exit 1; }
 
+# A maintainer can confirm a defect and claim it in the same breath. Grading the
+# sentiment first scores it as approval and never reaches the claim, so the claim
+# test must run before the tiering.
+want 914 SKIP "maintainer confirmed it and claimed it"
+want_reason 914 already
+
+# A non-maintainer claim is not an assignment, so the assignee test cannot see it.
+want 915 SKIP "outside contributor claimed it recently"
+want_reason 915 claimed
+
+# ...but a claim nobody acted on must not fence the issue off forever.
+want 916 KEEP "stale claim is abandoned, not competing"
+
+# Signal tiers. Association is the prerequisite; the comment text sets the grade.
+[ "$(field 903 maintainer_signal)" = "invites_pr" ] || { echo "FAIL #903 signal: got '$(field 903 maintainer_signal)'"; exit 1; }
+[ "$(field 911 maintainer_signal)" = "invites_pr" ] || { echo "FAIL #911 signal: got '$(field 911 maintainer_signal)'"; exit 1; }
+[ "$(field 919 maintainer_signal)" = "confirms" ]   || { echo "FAIL #919 signal: got '$(field 919 maintainer_signal)'"; exit 1; }
+[ "$(field 918 maintainer_signal)" = "neutral" ]    || { echo "FAIL #918 signal: got '$(field 918 maintainer_signal)'"; exit 1; }
+
+# A comment with no prose cannot be graded as prose. A bare link is a pointer,
+# not an endorsement, and no sentiment tier can read one.
+want 917 KEEP "bare link is not a defect signal"
+[ "$(field 917 maintainer_signal)" = "none" ] || { echo "FAIL #917 signal: got '$(field 917 maintainer_signal)'"; exit 1; }
+
+# A union-only lead still earns no signal, whatever they wrote.
+[ "$(field 904 maintainer_signal)" = "none" ] || { echo "FAIL #904 signal: got '$(field 904 maintainer_signal)'"; exit 1; }
+
+# The strongest signal is also the shortest one anyone writes. A prose floor
+# applied before the match would grade this as no signal at all.
+want 920 KEEP "terse invite"
+[ "$(field 920 maintainer_signal)" = "invites_pr" ] || { echo "FAIL #920 signal: got '$(field 920 maintainer_signal)'"; exit 1; }
+
+# "on our side" attached to the defect owns the bug; attached to the work claims
+# it. Only the second is a skip.
+want 921 KEEP "maintainer owns the defect without claiming the fix"
+[ "$(field 921 maintainer_signal)" = "confirms" ] || { echo "FAIL #921 signal: got '$(field 921 maintainer_signal)'"; exit 1; }
+
+# A comment can serialize without createdAt, and `null | fromdateiso8601` throws,
+# which would abort the batch and take every other issue with it.
+want 922 KEEP "undated comment does not abort the batch"
+
 echo "OK test_triage_filter.sh"