autopilot-issue-triage-scheduler

User-facing issue TRIAGE queue. This skill SELECTS issues and THROTTLES fan-out. It does not implement and does not review PRs.

Compose autopilot-issue-triage-worker one issue per slot. Never borrow slots from autopilot-issue-delivery-scheduler, autopilot-pr-review-scheduler, or autopilot-pr-triage-scheduler.

Activation card

activation_card: on. Before any queue read or spawn, emit this Enter card with every field filled. Missing field -> stop.

skill: autopilot-issue-triage-scheduler
skill_path: <resolved directory of this SKILL.md>
mode: run
subject: microsoft/apm
path: triage
intent: select issues and fan out triage workers
origin: unattended | actor-session
write: off
repo: microsoft/apm
fanout_limit: <positive integer>
invocation: agentic-workflow | actor-session

Rules:

• write is always off. This scheduler never comments, labels, assigns, or requests reviewers.
• write: on -> stop.
• origin fail-closed unknown -> unattended.
• invocation is the harness. Copilot App, local session, Cloud Agent, and Remote Agent are actor-session. agentic-workflow is only gh-aw / GitHub Actions. Do not copy origin into invocation.
• One queue. Do not nest another scheduler path.

After the run, emit this Exit receipt:

skill: autopilot-issue-triage-scheduler
subject: microsoft/apm
path: triage
write: off
queued: <integer>
spawned: <integer>
approved: n/a

Invocation

Works in local sessions, Copilot App automations, Cloud Agent, Remote Agent, and Agentic Workflows.

Resolve ORIGIN before any GitHub write. No assignment needed. This scheduler and its slots never assign and never request reviewers, in any ORIGIN.

INTENT is triage.

Ownership writes:

• Issue triage: none (no assignment needed)
• Code (accepted implementation): assign the implementing user
• PR review: request the reviewing user as reviewer, never assignee

Selection (same contract as Triage Panel AW)

Canonical owner: this skill's scripts/fetch_queue.py, scripts/triage_state.py, and assets/label-contract.json. The PR-triage scheduler ships an identical copy. Do not invent a second filter. The worker does not list the queue.

Modes (match .github/workflows/triage-panel.md):

• Named list or one issue number: explicit request (dispatch / label-event). triage/requested is the only request trigger. status/needs-triage is human decision state, not an event, and is not consumed.
• queue-all / no names: daily-style sweep.

fetch_queue.py owns list + eligibility (closed, locked, bot, empty, template-only; sweep-only spam). Sweep list excludes processing.read_reviewed (triage/recommended, status/triaged) at GitHub so already-advised open items are not downloaded. Do not comment on skips. triage_state.py owns completed-advice skip (processing.read_reviewed), at most two per author, oldest first, cap 10. Explicit requests bypass only spam and completed-advice.

Default (authenticated gh):

python <this-skill>/scripts/fetch_queue.py --kind issue --mode sweep --repo microsoft/apm > batch.json
python <this-skill>/scripts/triage_state.py < batch.json

Named issue: --mode dispatch --number N. If gh is unavailable, pass --records-json and --labels-json. A helper failure stops the run.

Do not implement. Do not open PRs. Do not fill delivery or review slots from this pool. Do not comment on issues. Do not add or remove labels. add_labels from triage_state.py is a plan for the worker, not a scheduler write.

Queue table (mandatory)

Before any spawn, emit the keep-set and the drop-set in plan.md and in the session report. Missing table, missing column, or blank rationale -> stop. Do not spawn.

Keep-set (items this run will schedule). One row per item. Do not truncate to FANOUT_LIMIT:

| number | kind | labels | rationale | slot |

Drop-set (considered, then not scheduled). slot is -:

| number | kind | labels | rationale | slot |

• kind: issue or pr
• labels: current GitHub labels, comma-separated; none if empty
• rationale: why queued or dropped (helper rule and labels)
• slot: 1-based spawn order, or - when dropped

Empty keep-set is success. Still emit the drop-set, or none. Completed-advice excluded at fetch is not a drop-set row. You are the sole table writer.

Fan-out

Load assets/fan-out-pool.md. Default FANOUT_LIMIT=2 concurrent slots. Isolated to this run. FANOUT_LIMIT is concurrency, not queue length. Drain the full selected list; when a slot returns, fill it with the next item.

Procedure

1. Probe scripts/fetch_queue.py, scripts/triage_state.py, and autopilot-issue-triage-worker on disk. Missing -> stop.
2. Build the queue with the helpers. Emit the mandatory queue table (labels + rationale) before any spawn.
3. Drain the selected list. Concurrent slots <= FANOUT_LIMIT. One issue per slot. Prefer a new session, else a sub-agent, else run the worker in this thread for that one issue, then the next. When a slot returns, dispatch the next queued issue. Do not stop because the pool was full. Each slot reads the complete comment history before advising.
4. Optionally present ONE consolidated triage digest. Wait if the caller asked for a checkpoint. Never treat silence as accept.
5. Print a final report from the table.

Hard nos

• Do not share this pool.
• Do not dispatch the same issue to two slots.
• Do not implement inside the scheduler thread.
• Do not comment, label, close, or assign. Workers own those writes even when summoned without this scheduler.
• Do not contradict CODEOWNERS.
• ASCII only.