rt-submission-readiness
Use when an author is one revision away from submission and needs a venue-parameterized go/no-go scorecard on the actual manuscript: fit, identifica…
它会碰到什么
这一栏是扫描器报的事实,不是结论。命中多不等于有毒(安全工具、规则库、示例脚本本来就会包含危险写法),命中少也不等于干净。它和你手上的凭据、文件、网络有什么关系,需要你自己看。
技能内容
Submission-Readiness Self-Check (rt-submission-readiness)
The mechanical/structural go/no-go that catches avoidable desk rejects before you hit
submit. Full rubric:
[shared-resources/submission-readiness/readiness-checklist.md](../../../shared-resources/submission-readiness/readiness-checklist.md).
It is the outward-facing analogue of the repo's internal tools/quality_scorecard.py.
When to trigger
- The draft is "done" and the author is about to submit to a named venue.
- Before paying a submission fee / starting the portal.
- After
rt-journal-matchselects the target and afterrt-execution-bridgehas run any
empirical checks the paper depends on.
Inputs to load
- The manuscript or submission packet, including front matter, main text, exhibits,
appendix, data/code statement, and cover letter if drafted.
- The target journal pack's
README.md, relevant skills, and
resources/official-source-map.md.
- Any executed analysis artifacts from
rt-execution-bridge: result handles, audit gaps,
robustness outputs, and replication/package notes.
What it does
Reads the target pack's skills + resources/official-source-map.md, inspects the actual
manuscript (cite §/table for every flag), and scores nine dimensions PASS / FLAG / FAIL:
- Fit & general interest · 2. Contribution clarity · 3. Identification / method ·
- Robustness & inference · 5. Exhibits self-contained · 6. Exposition (magnitude up front)
· 7. Venue mechanics (page limit / anonymization / fee / cover letter) · 8. Data & code /
reproducibility · 9. Pre-empted referee objections.
Go/no-go: any FAIL on Fit, Identification, or Venue-mechanics is a no-go (the classic
desk-reject triggers).
Decision contract
| Result | Meaning | Required next action |
|---|---|---|
| GO | No blocking FAIL and all venue mechanics are verified. | Run rt-simulated-referee for the substantive attack surface before submission. |
| CONDITIONAL GO | No fatal desk-reject trigger, but one or more FLAG items will draw reviewer pressure. | Fix or explicitly justify each FLAG; rerun this check on the changed sections. |
| NO-GO | Any FAIL on fit, identification/method, or venue mechanics, or an unverifiable required venue rule. | Route each blocking item to the owning target-pack skill and rerun before opening the portal. |
For UNKNOWN, name the missing evidence and the smallest artifact needed to resolve it. Do
not silently upgrade UNKNOWN to PASS.
Hard rules
- Read the manuscript; cite locations — no status without a §/table/page reference.
- Venue facts live from the source-map, never from memory.
- No false green — unverifiable dimension →
UNKNOWN, not PASS. - Map every FAIL/FLAG to the owning skill (and its execution bridge) so the author
knows exactly where to fix it.
- Separate mechanics from judgment — this skill catches preventable desk-reject risk;
use rt-simulated-referee for the substantive probability of surviving peer review.
Output format
【Venue】… (bar: pack skills + source-map, read live)
【Scorecard】1 Fit … PASS/FLAG/FAIL — note (§ cited) · … (dims 2–9)
【Go / No-go】GO / NO-GO — deciding dimension(s)
【Blocking gaps (FAIL)】ranked; each with the fix + owning skill
【Will-be-pushed (FLAG)】referee-anticipation list
【Unknowns】missing evidence required before a green light
【Desk-reject risk】low / medium / high — why
Next: if readiness passes, rt-simulated-referee for the substantive adversarial review.
想直接用这个技能?
本站把开放许可(MIT / Apache 等)的技能按仓库打包整理到网盘,点一下转存到你自己的网盘,不用一个个从 GitHub 拉。许可未声明的技能只给原始仓库链接,不打包。
它属于哪个仓库
Research-Toolkit-Skills/skills/rt-submission-readiness/SKILL.md