peer-review

Name: peer-review
Author: yha9806

24stars3forks

How to install peer-review?

yha9806/academic-writing-toolkit/peer-review

$npx -y skills add yha9806/academic-writing-toolkit --skill peer-review

Installs into the current project.

Summary

Review another author's manuscript, paper, thesis chapter, proposal, or preprint as an external reviewer. Use when asked to evaluate novelty, significance, gap-contribution fit, claim-evidence adequacy, methods, evaluation, overclaim risks, structure, writing, required revisions,

Files · 1

View on GitHub

SKILL.md

1	# /peer-review - External Manuscript Review
2
3	## Purpose
4
5	Produce a reviewer-style assessment of another author's work. This skill judges the manuscript as submitted, not as the author wishes it to be.
6
7	Use `/argument-governance` first when the review needs a formal gap-contribution or claim-evidence map.
8
9	## Codex-Only Baseline
10
11	Complete the review with Codex using the supplied manuscript, supplements, rubric, and source anchors. Do not require Gemini, gemini-agent, a second model, or a subagent. If another reviewer tool is available, record its output as optional advisory context only.
12
13	If `/argument-governance` is unavailable in the user's environment, manually extract the same spine in the review report: problem, gap, contribution, main claims, evidence, boundary language, and reviewer risks.
14
15	## Enhanced Advisory Mode
16
17	If the user explicitly provides or enables an API-key-backed model review, use it only after Codex has made a first-pass review from the supplied materials.
18
19	Rules:
20
21	- keep API keys in environment variables, never in review files
22	- send only user-approved manuscript or supplement excerpts
23	- label the external result as `advisory_external_review`
24	- compare it against Codex's local findings
25	- include only grounded, source-checkable points in the final review
26	- do not treat the external model as a reviewer consensus, venue decision, or factual verifier
27
28	## Core Rules
29
30	1. Review only the manuscript and explicitly supplied supplementary materials.
31	2. Do not use private project memory, prior chats, unstated author intent, or unpublished context as evidence.
32	3. Separate reviewer judgment from factual verification.
33	4. Do not rewrite the manuscript unless the user explicitly asks for revision suggestions.
34	5. Attribute concerns to manuscript locations, sections, figures, tables, claims, or missing evidence.
35	6. Be fair: identify strengths before high-level weaknesses, but do not soften required revisions.
36	7. Do not infer venue-specific acceptance thresholds unless the venue or rubric is supplied.
37	8. Do not treat an unavailable external review tool as a blocker.
38
39	## Workflow
40
41	### 1. Establish Review Scope
42
43	Identify:
44
45	- manuscript path or pasted text
46	- article type or intended venue, if provided
47	- supplied supplementary materials
48	- whether the user wants a formal review, desk-reject triage, or improvement-oriented review
49	- whether reference or evidence checking is in scope
50
51	### 2. Extract The Claimed Paper Spine
52
53	From the manuscript only, identify:
54
55	- central problem
56	- claimed gap
57	- stated contributions
58	- main claims
59	- evidence offered for each claim
60	- limitations and boundary language
61
62	If this cannot be extracted, report that as a review finding.
63
64	### 3. Evaluate With The Reviewer Rubric
65
66	Read `references/reviewer_rubric.md` when producing a full review. Cover:
67
68	- novelty
69	- significance
70	- gap-contribution fit
71	- method or evaluation validity
72	- evidence adequacy
73	- claim-evidence balance
74	- related-work coverage
75	- overclaim and generalisation risk
76	- reproducibility or transparency
77	- structure and clarity
78
79	### 4. Write The Review
80
81	Use `references/reviewer_report_template.md` for formal reviews. For shorter triage, preserve the same categories but compress the prose.
82
83	Give required revisions as actions that the author could take:
84
85	- add or verify evidence
86	- narrow a claim
87	- move a claim from result to limitation
88	- clarify contribution
89	- separate background from validation
90	- add missing baseline, ablation, provenance, or reference checks
91
92	### 5. Optional Structured Outputs
93
94	When requested, create:
95
96	- `reviews/peer_review_report.md`
97	- `reviews/reviewer_attack_matrix.csv`
98	- `reviews/claim_evidence_findings.csv`
99
100	These files are advisory review artifacts. They are not final evidence for the author's claims.
101
102	## Recommendation Vocabulary
103
104	Use one of:
105
106	- `accept`
107	- `minor_revision`
108	- `major_revision`
109	- `reject_resubmit`
110	- `reject`
111	- `no_recommendation`
112
113	Use `no_recommendation` when the manuscript, venue, or review scope is too incomplete.
114
115	## Output Pattern
116
117	```text
118	## Peer Review
119
120	### Summary
121
122	### Strengths
123
124	### Major Concerns
125
126	### Gap-Contribution Fit
127
128	### Claim-Evidence Adequacy
129
130	### Method / Evaluation Concerns
131
132	### Overclaim And Boundary Risks
133
134	### Minor Comments
135
136	### Required Revisions
137
138	### Recommendation
139	```
140
141	## Stop Conditions
142
143	Stop and report a blocker if:
144
145	- the manuscript or review target is missing
146	- the user asks for a review based on memory rather than supplied materials
147	- the requested factual verification requires sources that are not supplied and w

People who liked this also viewed

6 picks

Type

peer-review

24stars3forks

View on GitHub

How to install peer-review?

yha9806/academic-writing-toolkit/peer-review

$npx -y skills add yha9806/academic-writing-toolkit --skill peer-review

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# /peer-review - External Manuscript Review
2
3	## Purpose
4
5	Produce a reviewer-style assessment of another author's work. This skill judges the manuscript as submitted, not as the author wishes it to be.
6
7	Use `/argument-governance` first when the review needs a formal gap-contribution or claim-evidence map.
8
9	## Codex-Only Baseline
10
11	Complete the review with Codex using the supplied manuscript, supplements, rubric, and source anchors. Do not require Gemini, gemini-agent, a second model, or a subagent. If another reviewer tool is available, record its output as optional advisory context only.
12
13	If `/argument-governance` is unavailable in the user's environment, manually extract the same spine in the review report: problem, gap, contribution, main claims, evidence, boundary language, and reviewer risks.
14
15	## Enhanced Advisory Mode
16
17	If the user explicitly provides or enables an API-key-backed model review, use it only after Codex has made a first-pass review from the supplied materials.
18
19	Rules:
20
21	- keep API keys in environment variables, never in review files
22	- send only user-approved manuscript or supplement excerpts
23	- label the external result as `advisory_external_review`
24	- compare it against Codex's local findings
25	- include only grounded, source-checkable points in the final review
26	- do not treat the external model as a reviewer consensus, venue decision, or factual verifier
27
28	## Core Rules
29
30	1. Review only the manuscript and explicitly supplied supplementary materials.
31	2. Do not use private project memory, prior chats, unstated author intent, or unpublished context as evidence.
32	3. Separate reviewer judgment from factual verification.
33	4. Do not rewrite the manuscript unless the user explicitly asks for revision suggestions.
34	5. Attribute concerns to manuscript locations, sections, figures, tables, claims, or missing evidence.
35	6. Be fair: identify strengths before high-level weaknesses, but do not soften required revisions.
36	7. Do not infer venue-specific acceptance thresholds unless the venue or rubric is supplied.
37	8. Do not treat an unavailable external review tool as a blocker.
38
39	## Workflow
40
41	### 1. Establish Review Scope
42
43	Identify:
44
45	- manuscript path or pasted text
46	- article type or intended venue, if provided
47	- supplied supplementary materials
48	- whether the user wants a formal review, desk-reject triage, or improvement-oriented review
49	- whether reference or evidence checking is in scope
50
51	### 2. Extract The Claimed Paper Spine
52
53	From the manuscript only, identify:
54
55	- central problem
56	- claimed gap
57	- stated contributions
58	- main claims
59	- evidence offered for each claim
60	- limitations and boundary language
61
62	If this cannot be extracted, report that as a review finding.
63
64	### 3. Evaluate With The Reviewer Rubric
65
66	Read `references/reviewer_rubric.md` when producing a full review. Cover:
67
68	- novelty
69	- significance
70	- gap-contribution fit
71	- method or evaluation validity
72	- evidence adequacy
73	- claim-evidence balance
74	- related-work coverage
75	- overclaim and generalisation risk
76	- reproducibility or transparency
77	- structure and clarity
78
79	### 4. Write The Review
80
81	Use `references/reviewer_report_template.md` for formal reviews. For shorter triage, preserve the same categories but compress the prose.
82
83	Give required revisions as actions that the author could take:
84
85	- add or verify evidence
86	- narrow a claim
87	- move a claim from result to limitation
88	- clarify contribution
89	- separate background from validation
90	- add missing baseline, ablation, provenance, or reference checks
91
92	### 5. Optional Structured Outputs
93
94	When requested, create:
95
96	- `reviews/peer_review_report.md`
97	- `reviews/reviewer_attack_matrix.csv`
98	- `reviews/claim_evidence_findings.csv`
99
100	These files are advisory review artifacts. They are not final evidence for the author's claims.
101
102	## Recommendation Vocabulary
103
104	Use one of:
105
106	- `accept`
107	- `minor_revision`
108	- `major_revision`
109	- `reject_resubmit`
110	- `reject`
111	- `no_recommendation`
112
113	Use `no_recommendation` when the manuscript, venue, or review scope is too incomplete.
114
115	## Output Pattern
116
117	```text
118	## Peer Review
119
120	### Summary
121
122	### Strengths
123
124	### Major Concerns
125
126	### Gap-Contribution Fit
127
128	### Claim-Evidence Adequacy
129
130	### Method / Evaluation Concerns
131
132	### Overclaim And Boundary Risks
133
134	### Minor Comments
135
136	### Required Revisions
137
138	### Recommendation
139	```
140
141	## Stop Conditions
142
143	Stop and report a blocker if:
144
145	- the manuscript or review target is missing
146	- the user asks for a review based on memory rather than supplied materials
147	- the requested factual verification requires sources that are not supplied and w

People who liked this also viewed

6 picks

Type