All skills
imbad0202 avatar

/academic-paper-reviewer

@a3f6569

Multi-perspective academic paper review with dynamic reviewer personas. Runs a 5-seat, role-separated review panel (Journal-Fit Reviewer + 3 peer-review roles + Devil's Advocate) with field-specific expertise; role separation is not a claim of independent error processes. Supports full review, re-review (verification), quick assessment, methodology focus, Socratic guided, and calibration modes. Triggers on: review paper, peer review, manuscript review, referee report, review my paper, critique paper, simulate review, editorial review, calibrate reviewer, reviewer calibration, measure reviewer accuracy, 審查論文, 論文審查, 模擬審查, 同儕審查, 幫我審這篇, 以審查人角度評估, 審查者校準, 논문 심사, 동료 심사, 모의 심사, 심사자 관점에서 평가, 심사자 보정, revisar artículo, revisión entre pares, revisión de manuscrito, informe de árbitro, revisa mi artículo, criticar artículo, simular revisión, revisión editorial, calibrar revisor.

Use this Skill: https://skilld.dev/gh/imbad0202/academic-research-skills/academic-paper-reviewer

This session only. Nothing lands on disk.

templateseditorial_decision_template.md

≈2.8k tokens on demand. Your agent reads this file only when SKILL.md points to it.

Editorial Decision Template

This template is used by editorial_synthesizer_agent to produce the final Editorial Decision Package.


Template

# Editorial Decision

## Manuscript Information
- **Title**: [Paper title]
- **Manuscript ID**: [If available]
- **Submission Date**: [Submission date]
- **Decision Date**: [Decision date]
- **Review Round**: [Round N]

## Review Panel Provenance (#540/#740)

[`reviewer_full` letters only — other modes describe their own panel composition and omit this block. Build the artifact only from the exact EIC/R1/R2/R3/DA roster bound to the canonical `reviewer_full` contract, then validate the closed Schema 6 carrier against the artifact's exact raw bytes and deterministic replay according to `references/review_panel_provenance_protocol.md`; render its values without inference.]

- **Typed artifact**: [path or passport reference]
- **Artifact SHA-256**: [raw bytes `artifact_sha256`]
- **Panel ID**: [artifact `panel_id`]
- **Normalized manifest SHA-256**: [artifact `normalized_manifest_sha256`]
- **Execution topology SHA-256**: [artifact `execution_topology_sha256`]
- **Fresh-context scope**: `within_panel_attempt_only` [does not compare retries or prior rounds]

| Seat | Role ID | Actor type | Context ID | Peer outputs visible | Model family | Provider | Human reviewer ID |
|---|---|---|---|---|---|---|---|
| [artifact seat] | [value or `unknown`] | [value] | [value or `unknown`] | [value or `unknown`] | [value or `unknown`] | [value or `unknown`] | [value or `unknown`] |

| Provenance axis | Status (`true` / `false` / `unknown`) |
|---|---|
| Role-separated | [artifact `axes.role_separated`] |
| Within-panel invocation-context separation | [artifact `axes.fresh_context`] |
| Blind to peer outputs | [artifact `axes.blind_to_peer_outputs`] |
| Model-family distinct | [artifact `axes.model_family_distinct`] |
| Provider distinct | [artifact `axes.provider_distinct`] |
| Human-reviewer distinct | [artifact `axes.human_distinct`] |

- **Binary independence claim**: Not computed. Persona or role diversity proves only `role_separated`; do not relabel the panel as independent.
- **Correlated-error disclosure**: [Render artifact `correlated_error_disclosure.text` verbatim when `required: true`; otherwise write "Not required by the model-family axis; no conclusion is implied for the other axes."]

---

## Decision *

### [Accept / Minor Revision / Major Revision / Reject]

[If Reject, indicate subtype: Out of Scope / Fundamental Flaw / Insufficient Contribution / Premature / Resubmit Encouraged]

---

## Blocking Issues * (0–3, immutable source order)

<!-- #574 E7: the 0-3 issues that currently BLOCK acceptance, in immutable
     roadmap source order,
     each with its evidence anchor and the roadmap item that resolves it, so the
     author does not have to synthesize the blockers across five long reports.
     ZERO rows is valid for a genuine Accept — never manufacture blockers to
     fill the section. -->

| Transport ref | Blocking issue | Source reviewer(s) | Evidence anchor | Resolving roadmap item |
|---------------|----------------|--------------------|-----------------|------------------------|
| R1 | [Issue] | [EIC/R1/R2/R3/DA] | [typed — `<type>: <locator>`, transported from the finding (#574 A2)] | [REV-n] |

---

## Reviewer Summary

| Reviewer | Role | Recommendation | Confidence |
|----------|------|---------------|------------|
| Journal-Fit Reviewer | [Senior-editor or associate-editor identity] | [Accept/Minor/Major/Reject] | [1-5] |
| Reviewer 1 | [Methodology expert identity] | [Accept/Minor/Major/Reject] | [1-5] |
| Reviewer 2 | [Domain expert identity] | [Accept/Minor/Major/Reject] | [1-5] |
| Reviewer 3 | [Cross-disciplinary expert identity] | [Accept/Minor/Major/Reject] | [1-5] |
| Devil's Advocate | Fixed adversarial seat | N/A — findings only | N/A — per-finding only |

---

## Consensus Analysis *

### Points of Agreement (Consensus)

**[CONSENSUS-4]** (All 4 non-DA scoring reviewers agree):
1. [Consensus content — cite relevant passages from each reviewer's report]
2. [...]

**[CONSENSUS-3]** (3/4 non-DA scoring reviewers agree, the 4th **silent**):
1. [Consensus content — indicate which 3 agree and name the silent 4th. If the 4th *disputes* the sub-claim rather than being silent, it is a SPLIT (see Points of Disagreement), not a CONSENSUS-3.]
2. [...]

### Points of Disagreement

**Disagreement 1: [Issue name]**
- **R[X] view**: [Specific viewpoint, citing report]
- **R[Y] view**: [Specific viewpoint, citing report]
- **Disagreement type**: [Perspective difference / Severity disagreement / Existence disagreement / Direction disagreement]
- **Editor's Resolution**: [Arbitration result]
- **Resolution Rationale**: [Arbitration rationale — based on evidence/expertise/unresolved-dissent principle (#574 B1)]

**Disagreement 2: [Issue name]**
- [Same format as above]

---

## Decision Rationale *

[200-300 words explaining the basis for this decision]

Requirements:
- Cite specific reviewer opinions
- Explain how disagreements were resolved
- Explain why this decision was chosen rather than a more or less strict one
- If Reject, explain why revision also cannot salvage it

---

## Required Revisions * (Must Fix)

[Only needed for Minor Revision and Major Revision]

| Transport ref | Revision Item | Sub-Claim(s) | Severity | Evidence Anchor | Confidence | Source Reviewer | Obligation class | Cost scope | Bounded consequence |
|---|--------------|--------------|----------|-----------------|------------|----------------|------------------|------------|---------------------|
| R1 | [Description] | [SC-n] | [transported: critical/major (+ fallback tag if any)] | [`<type>: <locator>`] | [n — basis] | [EIC/R1/R2/R3] | must_fix | [sentence/section/re-analysis/new-data/other + locator] | [closed code + target] |
| R2 | [Description] | [SC-n] | [transported] | [transported] | [transported] | [Source] | must_fix | [typed scope] | [closed consequence] |
| R3 | [Description] | [SC-n] | [transported] | [transported] | [transported] | [Source] | must_fix | [typed scope] | [closed consequence] |
...

The `Sub-Claim(s)` column carries the Step 1b `sub_claim_id`(s) the item traces to (e.g. `SC-1`); a DA-CRITICAL or non-decomposed item uses `—`.

### Required Item Details

> **Ordinal contract (#576/#670):** `R<n>` is a transport reference, never a work rank. Numbering follows the immutable Revision Roadmap's deterministic source-traceability order filtered to `obligation_class == must_fix`; it never reads author-selected display order or author triage. Required blocks are exactly `R1..Rn` with no gaps, duplicates, or extras. The **Acceptance criteria** field stays a SINGLE-LINE bullet (`- **Acceptance criteria**: <text>`) for `scripts/check_re_review_synthesis.py`.

**R1: [Title]**
- **Problem**: [Specific description]
- **Source**: [Which reviewer raised it, citing report passage]
- **Requirement**: [Specifically how to fix it]
- **Acceptance criteria**: [How to confirm the issue is resolved after fixing]

**R2: [Title]**
- [Same format as above]

---

## Suggested Revisions (Should Fix)

| Transport ref | Revision Item | Sub-Claim(s) | Severity | Evidence Anchor | Confidence | Source Reviewer | Obligation class | Cost scope | Bounded consequence |
|---|--------------|--------------|----------|-----------------|------------|----------------|------------------|------------|---------------------|
| S1 | [Description] | [SC-n] | [transported] | [transported] | [transported] | [Source] | should_fix | [typed scope] | [closed consequence + target] |
| S2 | [Description] | [SC-n] | [transported] | [transported] | [transported] | [Source] | consider | [typed scope] | [closed consequence + target] |
...

---

## Revision Roadmap *

### Source-traceability checklist

> Keep this in immutable source order. Do not suggest a work order. The author
> chooses `will_address`, `wont_address`, or `not_on_point` later in the
> separate author-adjudication checkpoint.

- [ ] R1 — obligation `must_fix`: [Task description]
- [ ] R2 — obligation `must_fix`: [Task description]
- [ ] S1 — obligation `should_fix`: [Task description]
- [ ] S2 — obligation `consider`: [Task description]

---

## Journal-Supplied Deadline (Optional Transport)

- **Exact deadline from source letter**: [verbatim date, or `NOT PROVIDED`]
- Do not infer a deadline, duration, or work estimate.

---

## Response Letter Instructions

Please use the format in `templates/revision_response_template.md` to respond to every reviewer comment item by item.

**Must include**:
1. Response and revision description for each Required Revision
2. Response for each Suggested Revision (adopted or reason for not adopting)
3. Change markup (mark all changes in the revised manuscript with color or track changes)
4. Cross-reference table of new page numbers/paragraphs

---

## Closing

[Formal closing, adjusting tone based on decision type]

### Accept Version
We are pleased to accept your manuscript for publication in [Journal Name]. [If applicable, include minor suggestions]

### Minor Revision Version
We invite you to submit a revised version of your manuscript, addressing the points raised by the reviewers. We look forward to receiving your revision within [deadline].

### Major Revision Version
We encourage you to carefully consider the reviewers' comments and submit a substantially revised manuscript. Please note that the revised manuscript will undergo another round of review.

### Reject Version
After careful consideration, we are unable to accept your manuscript for publication in [Journal Name]. We appreciate the effort you have put into this work and hope the reviewers' comments will be helpful for future development of this research.

[If appropriate, recommend alternative journals]

---

## Appendix: Full Reviewer Reports

[Attach all 5 complete reviewer reports — four card-backed scoring reports plus the fixed Devil's Advocate — for the author's reference]

---

## Attachment: Acronym Check (advisory, #849)

[Leave this section out. The dispatching session appends it after the decision is final (`SKILL.md` § Acronym check attachment).]

Format Guidelines

Revision Roadmap Design Principles

  1. Actionability: Every item is a concrete task, not an abstract suggestion
  2. Traceability: Every item can be traced back to specific reviewer comments
  3. Independent fields: Transport reviewer severity unchanged; record the editorial obligation, typed cost surface, and bounded consequence separately
  4. No work ranking: Immutable rows use source-traceability order; only the author may select a presentation view or decide what to address
  5. Exact scope: Every item proposes exact block/operation targets; proposal is not write authority
  6. Compatibility: Emit the closed revision-roadmap/1.0 machine artifact for academic-paper revision mode

Finding severity is the Schema 6 enum (critical / major / minor), transported from the reviewer cards (#574 A3). obligation_class is a separate editorial gate and is never derived as a work rank from severity.

Source: SKILL.md on GitHub

1 warning6d5 checks · Risk SAFE
  • Gen Agent Trust Hub6d

    The skill is a multi-agent framework for academic paper review. It is well-architected with significant security defenses against prompt injection from the manuscripts it processes. The primary risk is the large attack surface provided by untrusted input data, though this is mitigated by explicit boundary instructions.

  • Socket6d

    No alerts

  • Snyk6d

    Risk: MEDIUM · 1 issue

  • Runlayer6mo

    18 files scanned · No issues

  • ZeroLeaks5mo

    Score: 93/100 · 2 sections analyzed

Signed by skilld at a3f6569. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 34 minutes ago.

Activeupdated last week
Other metadata
metadata
{
  "version": "1.11.1",
  "last_updated": "2026-08-15",
  "status": "active",
  "data_access_level": "raw",
  "task_type": "open-ended",
  "related_skills": [
    "academic-paper",
    "academic-pipeline"
  ]
}

README badge

README badge for imbad0202/academic-research-skills/academic-paper-reviewer