---

name: rt-workflow description: Use to sequence a manuscript across venues — which cross-journal toolkit skill to invoke next, from picking a target journal through execution, submission-readiness, simulated review, rebuttal, and the replication package. Routes; it does not replace the per-journal depth packs or the specialized toolkit skills.

Cross-Journal Workflow Router (rt-workflow)

The author-side loop that runs across journals, complementing the per-journal depth packs (which answer "how do I clear THIS venue's bar?"). Each step is backed by a canonical capability doc in shared-resources/.

When to trigger

• You have a result/draft and don't yet know the venue, or just got a reject and need the next target.
• You are mid-project and want the next cross-journal step.
• You want the end-to-end sequence: pick → execute → ready → rehearse → submit → rebut → replicate.

The loop (and the skill that owns each step)

0. paper-profile.yml — write the five signals down once. Every skill below reads it instead of re-deriving them from the manuscript and disagreeing with itself.
1. rt-journal-match — profile the paper → ranked reach/match/safe shortlist + resubmission ladder.
2. rt-ladder-ev — cost the ladder in months and in probability of placing, when there is a clock or two candidate orders.
3. rt-venue-integrity — when the chosen venue is outside the index, verify it against primary sources before submitting.
4. rt-venue-reframe — when the target changes (a reject, or a switch), diff the manuscript from the old venue's framing to the new one.
5. rt-execution-bridge — run the analysis through the StatsPAI / Stata MCP tools (estimate + audit), not just advise it.
6. rt-submission-readiness — go/no-go scorecard on the manuscript against the target venue's bar.
7. rt-desk-reject-risk — score the draft against the target's own documented desk-reject triggers before uploading.
8. rt-simulated-referee — rehearse a calibrated AE + referee panel; surface the decisive objections before a real referee does.
9. (fix) — route each objection back to the target pack's skill + the execution bridge.
10. rt-response-to-referees — after an R&R, draft the point-by-point reply + revision plan (editor first).
11. rt-replication-package — assemble + validate the Data-Editor package against the venue's data-and-code policy.

Routing table

Symptom	Go to
"Where should I send this?" / "Rejected — what next?"	rt-journal-match
"Top journal first, or start one rung down?" / a tenure or job-market clock	rt-ladder-ev
"Is this journal real?" / an unfamiliar venue, a solicitation email, an index requirement	rt-venue-integrity
"What has to change to send it to X instead?"	rt-venue-reframe
"I need to actually run the DiD/IV/RDD/SCM/DML, not just plan it"	rt-execution-bridge
"Is it ready to submit?"	rt-submission-readiness
"Will the editor desk-reject this?"	rt-desk-reject-risk
"What will referees attack?"	rt-simulated-referee
"I have an R&R to answer"	rt-response-to-referees
"The Data Editor needs a replication package"	rt-replication-package

How it relates to the per-journal packs

This router sequences the cross-journal steps; the venue bar and all live facts (fees, limits, acceptance, data policy, house style) come from the chosen pack's own skills and resources/official-source-map.md. Always defer venue-specific judgment to that pack — this toolkit picks the venue, runs the analysis, and rehearses the gauntlet.

Anti-patterns

• Submitting before rt-submission-readiness clears the mechanical bar.
• Picking a venue without reading its live source-map facts (rt-journal-match hard rule).
• Treating the simulated referee as a verdict rather than a rehearsal.

Output format

【Current step】pick / execute / ready / rehearse / respond / replicate
【Next skill】rt-...
【Why】one line
【Venue bar source】<target pack>/resources/official-source-map.md + skills