Merge pull request 'Add a human-merge label that keeps autopilot from merging' (#126) from ai/autopilot-human-merge into main

autopilot skips both human checkpoints: approval before the build and the
merge. Scheduled update issues in homelab want the hands-off build but a human
merge, since merging there deploys. With human-merge also on the issue, @pm is
told to hand back at phase 3 and publish.sh refuses MERGE_PR.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

.gitea/workflows/agent.yml

@@ -268,6 +268,7 @@ jobs:
           HAS_IMAGES: ${{ steps.imgs.outputs.has_images }}
           BRANCH: ${{ steps.prep.outputs.branch }}
           AUTOPILOT: ${{ steps.prep.outputs.autopilot }}   # 'true' when the issue carries the `autopilot` label
+          AUTOMERGE: ${{ steps.prep.outputs.automerge }}   # 'false' when `human-merge` is also set
           NUM: ${{ github.event.issue.number }}
           TITLE: ${{ github.event.issue.title }}
           IBODY: ${{ github.event.issue.body }}
@@ -305,6 +306,7 @@ jobs:
           NEW: ${{ steps.prep.outputs.new }}
           IS_PR: ${{ github.event.issue.pull_request }}   # set when this run is on a PR thread
           AUTOPILOT: ${{ steps.prep.outputs.autopilot }}   # 'true' when the origin issue carries `autopilot`
+          AUTOMERGE: ${{ steps.prep.outputs.automerge }}   # 'false' when `human-merge` is also set
           ISSNUM: ${{ steps.prep.outputs.issnum }}          # origin issue number (resolved from branch on PR threads)
         run: bash "$SCRIPTS/publish.sh"
 

.gitea/workflows/scripts/publish.sh

@@ -4,7 +4,7 @@
 # Required env (provided by the workflow step):
 #   GT TOKEN_PM TOKEN_SENIOR TOKEN_JUNIOR TOKEN_LEAD TOKEN_QA
 #   NAME MODE NUM TITLE BRANCH NEW GITHUB_SERVER_URL GITHUB_REPOSITORY
-#   IS_PR AUTOPILOT ISSNUM  (autopilot: @qa label-gated merge/halt + auto-trigger @qa on a fresh PR)
+#   IS_PR AUTOPILOT AUTOMERGE ISSNUM  (autopilot: @qa label-gated merge/halt + auto-trigger @qa on a fresh PR)
 set +e   # publish is best-effort: a grep-no-match / curl non-zero must NOT kill the step
 # Post/PR as the agent's OWN Gitea user when its token is configured; else the built-in bot.
 case "$NAME" in
@@ -222,9 +222,10 @@ if [ "$MODE" != "pr" ]; then
   fi
 
   # --- @pm autopilot merge: @pm is the ONLY agent that merges, and ONLY under the autopilot label ---
+  # (and never when `human-merge` is also set — AUTOMERGE is false then).
   # (@qa never merges — it approves and hands back here.) Merge with the PAT (TTOK), not the built-in
   # token, so the push to main fires the deploy. TOKEN_PM must carry write:repository.
-  if [ "$NAME" = "pm" ] && [ "$AUTOPILOT" = "true" ] && grep -qiE '^[[:space:]]*MERGE_PR[[:space:]]*$' /tmp/agent_out.md; then
+  if [ "$NAME" = "pm" ] && [ "$AUTOMERGE" = "true" ] && grep -qiE '^[[:space:]]*MERGE_PR[[:space:]]*$' /tmp/agent_out.md; then
     PRS=$(resolve_prs); PRN=${PRS##* }; CNT=$(echo "$PRS" | wc -w)
     if [ -z "$PRN" ]; then
       echo "MERGE_PR but no open PR found for issue #$ISSN"

.gitea/workflows/scripts/route.sh

@@ -44,7 +44,7 @@ if [ -z "$name" ]; then
     # mode=skip so every later step no-ops — a red run for a non-agent comment is just noise.
     echo "no known agent mentioned (word-boundary) — skipping run"
     { echo "name=none"; echo "model=none"; echo "fallback="; echo "vision=false"; echo "mode=skip"; echo "skills=[]";
-      echo "branch=main"; echo "new=false"; echo "autopilot=false"; echo "issnum=$NUM"; } >> "$GITHUB_OUTPUT"
+      echo "branch=main"; echo "new=false"; echo "autopilot=false"; echo "automerge=false"; echo "issnum=$NUM"; } >> "$GITHUB_OUTPUT"
     exit 0
   fi
 fi
@@ -121,10 +121,12 @@ fi
 issnum="$NUM"
 case "$IS_PR" in ?*) issnum=$(printf '%s' "$branch_ref" | sed -nE 's,^ai/issue-([0-9]+).*,\1,p');; esac
 [ -z "$issnum" ] && issnum="$NUM"
-autopilot=false
-if curl -sS -H "Authorization: token $GT" "$API/issues/$issnum/labels" 2>/dev/null \
-     | jq -e 'any(.[]?; .name=="autopilot")' >/dev/null 2>&1; then
+# `human-merge` alongside it keeps the hands-off build but leaves the merge to a human.
+labels=$(curl -sS -H "Authorization: token $GT" "$API/issues/$issnum/labels" 2>/dev/null)
+autopilot=false; automerge=false
+if printf '%s' "$labels" | jq -e 'any(.[]?; .name=="autopilot")' >/dev/null 2>&1; then
   autopilot=true
+  printf '%s' "$labels" | jq -e 'any(.[]?; .name=="human-merge")' >/dev/null 2>&1 || automerge=true
 fi
-echo "autopilot (autopilot label on #$issnum)=$autopilot"
-{ echo "autopilot=$autopilot"; echo "issnum=$issnum"; } >> "$GITHUB_OUTPUT"
+echo "autopilot (autopilot label on #$issnum)=$autopilot automerge=$automerge"
+{ echo "autopilot=$autopilot"; echo "automerge=$automerge"; echo "issnum=$issnum"; } >> "$GITHUB_OUTPUT"

.gitea/workflows/scripts/run-agent.sh

@@ -3,7 +3,7 @@
 # plain-text reply (/tmp/agent_out.md) plus the raw event stream (/tmp/events.jsonl).
 #
 # Required env (provided by the workflow step):
-#   XAI_API_KEY SELF_TOKEN NAME MODEL VISION MODE HAS_IMAGES BRANCH AUTOPILOT NUM TITLE
+#   XAI_API_KEY SELF_TOKEN NAME MODEL VISION MODE HAS_IMAGES BRANCH AUTOPILOT AUTOMERGE NUM TITLE
 #   IBODY CMT
 #   FILES   (the opencode -f image flags, from the imgs step output)
 #   AUTOPILOT is 'true' when the issue carries the `autopilot` label (label-gated autopilot mode).
@@ -74,14 +74,21 @@ if [ "$MODE" = "comment" ]; then
    auto-start any dev — the maintainer @mentions an agent on each sub-issue when ready."
     if [ "$AUTOPILOT" = "true" ]; then
       ACTION="$ACTION
-   AUTOPILOT MODE IS ACTIVE (this issue carries the 'autopilot' label) — it changes exactly TWO
-   things for you; everything else above is unchanged:
+   AUTOPILOT MODE IS ACTIVE (this issue carries the 'autopilot' label) — it changes only the
+   phases below; everything else above is unchanged:
    - PHASE 1: do NOT ask '@ffaerber ready to build?'. Present your SHORT plan AND end with a
      'DELEGATE: @<dev>' line in the SAME reply. Only skip delegating (and ask @ffaerber) if the task
-     is genuinely ambiguous or unsafe.
+     is genuinely ambiguous or unsafe."
+      if [ "$AUTOMERGE" = "true" ]; then
+        ACTION="$ACTION
    - PHASE 3: do NOT ask the human to merge. When @qa has approved, end your reply with EXACTLY
      'MERGE_PR' — the automation merges the PR and closes this issue. You are the ONLY agent that
      merges, and only here."
+      else
+        ACTION="$ACTION
+   - PHASE 3 is UNCHANGED: this issue also carries 'human-merge', so when @qa has approved, tell
+     @ffaerber the PR is ready to merge. Do NOT emit 'MERGE_PR'."
+      fi
     fi
   fi
   if [ "$NAME" = "qa" ]; then

AGENTS.md

@@ -29,6 +29,8 @@ An issue labeled **`autopilot`** runs without the usual human checkpoints:
 - **Kill switch:** remove the `autopilot` label at any time. The label is re-read fresh at the
   start of every run, so the next agent turn reverts to normal human-approval behavior. `@qa` also
   removes the label itself whenever it halts on a bug or a failed merge.
+- **`human-merge` label** (together with `autopilot`): everything above except the merge — after
+  `@qa` approves, `@pm` tells the creator the PR is ready and a human merges.
 No label (the default) = today's behavior, unchanged.
 
 ## Branches & pull requests

README.md

@@ -31,7 +31,8 @@ deploy-host SSH skill, live in the consuming repo under `.gitea/agent-skills/`
 4. `@qa` reviews **on the PR** — either recommendations + `BOUNCE: @dev` (dev fixes → `@qa`
    re-verifies, direct loop, max 3 rounds) or `APPROVE`.
 5. On approval `@qa` posts the verdict **on the issue** → `@pm` tells the creator *"ready to merge"*
-   and a **human merges** — or, with the `autopilot` label, `@pm` merges and closes the issue itself.
+   and a **human merges** — or, with the `autopilot` label, `@pm` merges and closes the issue itself
+   (unless the issue also carries `human-merge`, which keeps the merge with a human).
 
 `@pm` is the only agent that ever merges, and only under the `autopilot` label (its kill switch:
 remove the label mid-flight and the next step reverts to human control).