self-review

Name: self-review
Author: yha9806

24stars3forks

How to install self-review?

yha9806/academic-writing-toolkit/self-review

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

Installs into the current project.

Summary

Review the user's own manuscript, paper, thesis chapter, rebuttal, or release packet with clean-room anti-contamination controls. Use when asked for self-review, internal review, pre-submission review, readiness check, reviewer simulation on own work, or claim-evidence self-audit

Files · 1

View on GitHub

SKILL.md

1	# /self-review - Clean-Room Manuscript Self-Review
2
3	## Purpose
4
5	Audit the user's own work without letting memory, prior chats, unstated project knowledge, or the model's background knowledge become evidence.
6
7	The governing rule is:
8
9	```text
10	self-review truth = explicit review packet + source anchor
11	```
12
13	Use `/argument-governance` first when the manuscript needs a formal intent, contribution, claim, and evidence map.
14
15	## Codex-Only Baseline
16
17	Complete self-review with Codex, the review manifest, allowed sources, and the bundled packet checker. Do not require Gemini, gemini-agent, a second model, or a subagent. If an external review is available, keep it in `Reviewer-risk inference` or advisory notes and never use it as source support.
18
19	If `/argument-governance` is unavailable, manually extract the same clean-room argument spine from manifest-listed sources only.
20
21	## Enhanced Advisory Mode
22
23	If the manifest and the user explicitly allow an API-key-backed advisory review, Codex may run or incorporate a second-model pass after the clean-room self-review packet is valid.
24
25	Rules:
26
27	- the base clean-room review must be possible without the external call
28	- API keys must be read from environment variables only
29	- the manifest may name `api_key_env_var`, but must never store the key value
30	- only manifest-approved source subsets may be sent externally
31	- external findings must be placed under `Reviewer-risk inference` or advisory notes
32	- external findings must be re-grounded against allowed sources before becoming revision actions
33	- unsupported external comments stay unsupported
34
35	## Core Rules
36
37	1. Use only files listed in the review manifest.
38	2. Treat prior chat memory, unstated project assumptions, model background knowledge, and unlisted notes as forbidden evidence.
39	3. Split every finding into `Supported by packet`, `Not supported by packet`, or `Reviewer-risk inference`.
40	4. Every supported finding must include a source anchor.
41	5. Do not repair missing evidence by remembering earlier conversations.
42	6. Do not treat generated reviews, agent drafts, or reviewer simulations as final evidence.
43	7. Do not edit the manuscript until the user approves specific revision actions.
44	8. Do not treat an unavailable external review tool as a blocker.
45
46	## Required Packet
47
48	The preferred layout is:
49
50	```text
51	review_packet/
52	review_manifest.yaml
53	manuscript.md or manuscript.pdf
54	references.bib
55	evidence/
56	figures/
57	tables/
58	claims/
59	```
60
61	Read `references/clean_room_protocol.md` before reviewing. Read `references/self_review_packet_schema.md` before creating or validating a packet.
62
63	## Workflow
64
65	### 1. Validate The Clean-Room Packet
66
67	Resolve the bundled helper at `scripts/check_self_review_packet.py` relative to this `SKILL.md`, then run:
68
69	```bash
70	python3 {skill_dir}/scripts/check_self_review_packet.py review_packet --json
71	```
72
73	If the packet is missing or invalid, report the issue before reviewing.
74
75	### 2. Build The Source-Bounded Reading List
76
77	Read only manifest-listed files. If a needed file is not listed, ask whether to add it to the manifest or mark the issue as unsupported.
78
79	### 3. Extract The Argument Spine
80
81	From the packet only, extract:
82
83	- stated intent
84	- named gap
85	- contributions
86	- main claims
87	- evidence anchors
88	- limitations
89	- reviewer-risk areas
90
91	### 4. Run Self-Review Checks
92
93	Check:
94
95	- gap-contribution alignment
96	- claim hierarchy
97	- claim-evidence fit
98	- evidence balance
99	- unsupported or overextended claims
100	- missing limitations
101	- reviewer attacks with weak defenses
102	- internal consistency and submission blockers
103
104	### 5. Write A Clean-Room Report
105
106	The report must separate:
107
108	- `Supported by packet`
109	- `Not supported by packet`
110	- `Reviewer-risk inference`
111
112	Do not merge these categories.
113
114	## Output Pattern
115
116	```text
117	## Clean-Room Self-Review
118
119	### Packet Boundary
120	- Manifest:
121	- Allowed sources:
122	- Forbidden sources:
123
124	### Supported By Packet
125	\| Finding \| Source anchor \| Severity \| Action \|
126
127	### Not Supported By Packet
128	\| Claim or need \| Missing source \| Risk \| Action \|
129
130	### Reviewer-Risk Inference
131	\| Risk \| Basis in packet \| Why it matters \| Action \|
132
133	### Revision Actions
134	\| Priority \| Action \| Requires new evidence? \|
135	```
136
137	## Stop Conditions
138
139	Stop and report a blocker if:
140
141	- no review manifest is provided
142	- the user asks to rely on previous chat or memory for evidence
143	- a central claim needs a source not listed in the manifest
144	- the packet validator reports missing allowed files
145	- the user asks to mark unsupported claims as supported

People who liked this also viewed

6 picks

Type

self-review

24stars3forks

View on GitHub

How to install self-review?

yha9806/academic-writing-toolkit/self-review

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

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# /self-review - Clean-Room Manuscript Self-Review
2
3	## Purpose
4
5	Audit the user's own work without letting memory, prior chats, unstated project knowledge, or the model's background knowledge become evidence.
6
7	The governing rule is:
8
9	```text
10	self-review truth = explicit review packet + source anchor
11	```
12
13	Use `/argument-governance` first when the manuscript needs a formal intent, contribution, claim, and evidence map.
14
15	## Codex-Only Baseline
16
17	Complete self-review with Codex, the review manifest, allowed sources, and the bundled packet checker. Do not require Gemini, gemini-agent, a second model, or a subagent. If an external review is available, keep it in `Reviewer-risk inference` or advisory notes and never use it as source support.
18
19	If `/argument-governance` is unavailable, manually extract the same clean-room argument spine from manifest-listed sources only.
20
21	## Enhanced Advisory Mode
22
23	If the manifest and the user explicitly allow an API-key-backed advisory review, Codex may run or incorporate a second-model pass after the clean-room self-review packet is valid.
24
25	Rules:
26
27	- the base clean-room review must be possible without the external call
28	- API keys must be read from environment variables only
29	- the manifest may name `api_key_env_var`, but must never store the key value
30	- only manifest-approved source subsets may be sent externally
31	- external findings must be placed under `Reviewer-risk inference` or advisory notes
32	- external findings must be re-grounded against allowed sources before becoming revision actions
33	- unsupported external comments stay unsupported
34
35	## Core Rules
36
37	1. Use only files listed in the review manifest.
38	2. Treat prior chat memory, unstated project assumptions, model background knowledge, and unlisted notes as forbidden evidence.
39	3. Split every finding into `Supported by packet`, `Not supported by packet`, or `Reviewer-risk inference`.
40	4. Every supported finding must include a source anchor.
41	5. Do not repair missing evidence by remembering earlier conversations.
42	6. Do not treat generated reviews, agent drafts, or reviewer simulations as final evidence.
43	7. Do not edit the manuscript until the user approves specific revision actions.
44	8. Do not treat an unavailable external review tool as a blocker.
45
46	## Required Packet
47
48	The preferred layout is:
49
50	```text
51	review_packet/
52	review_manifest.yaml
53	manuscript.md or manuscript.pdf
54	references.bib
55	evidence/
56	figures/
57	tables/
58	claims/
59	```
60
61	Read `references/clean_room_protocol.md` before reviewing. Read `references/self_review_packet_schema.md` before creating or validating a packet.
62
63	## Workflow
64
65	### 1. Validate The Clean-Room Packet
66
67	Resolve the bundled helper at `scripts/check_self_review_packet.py` relative to this `SKILL.md`, then run:
68
69	```bash
70	python3 {skill_dir}/scripts/check_self_review_packet.py review_packet --json
71	```
72
73	If the packet is missing or invalid, report the issue before reviewing.
74
75	### 2. Build The Source-Bounded Reading List
76
77	Read only manifest-listed files. If a needed file is not listed, ask whether to add it to the manifest or mark the issue as unsupported.
78
79	### 3. Extract The Argument Spine
80
81	From the packet only, extract:
82
83	- stated intent
84	- named gap
85	- contributions
86	- main claims
87	- evidence anchors
88	- limitations
89	- reviewer-risk areas
90
91	### 4. Run Self-Review Checks
92
93	Check:
94
95	- gap-contribution alignment
96	- claim hierarchy
97	- claim-evidence fit
98	- evidence balance
99	- unsupported or overextended claims
100	- missing limitations
101	- reviewer attacks with weak defenses
102	- internal consistency and submission blockers
103
104	### 5. Write A Clean-Room Report
105
106	The report must separate:
107
108	- `Supported by packet`
109	- `Not supported by packet`
110	- `Reviewer-risk inference`
111
112	Do not merge these categories.
113
114	## Output Pattern
115
116	```text
117	## Clean-Room Self-Review
118
119	### Packet Boundary
120	- Manifest:
121	- Allowed sources:
122	- Forbidden sources:
123
124	### Supported By Packet
125	\| Finding \| Source anchor \| Severity \| Action \|
126
127	### Not Supported By Packet
128	\| Claim or need \| Missing source \| Risk \| Action \|
129
130	### Reviewer-Risk Inference
131	\| Risk \| Basis in packet \| Why it matters \| Action \|
132
133	### Revision Actions
134	\| Priority \| Action \| Requires new evidence? \|
135	```
136
137	## Stop Conditions
138
139	Stop and report a blocker if:
140
141	- no review manifest is provided
142	- the user asks to rely on previous chat or memory for evidence
143	- a central claim needs a source not listed in the manifest
144	- the packet validator reports missing allowed files
145	- the user asks to mark unsupported claims as supported

People who liked this also viewed

6 picks

Type