argument-governance

Name: argument-governance
Author: yha9806

24stars3forks

How to install argument-governance?

yha9806/academic-writing-toolkit/argument-governance

$npx -y skills add yha9806/academic-writing-toolkit --skill argument-governance

Installs into the current project.

Summary

Build and audit the manuscript argument system: intent, gap-contribution alignment, hierarchical claims, evidence balance, limitations, and reviewer attack surfaces. Use when a paper, thesis chapter, review article, or rebuttal needs explicit intent registers, contribution chains

Files · 1

View on GitHub

SKILL.md

1	# /argument-governance - Argument System Governance
2
3	## Purpose
4
5	Make the paper's argument inspectable as a system:
6
7	```text
8	intent -> gap -> contribution -> claim hierarchy -> evidence -> boundary -> reviewer risk
9	```
10
11	Use this before major manuscript revision, before `/peer-review`, before `/self-review`, or after `/manuscript-reframe` when the draft needs stronger gap-contribution and claim-evidence control.
12
13	## Codex-Only Baseline
14
15	Complete this workflow with Codex, local files, and the bundled checker. External reviewers, subagents, Gemini, or other model calls are optional advisory inputs only; do not require them and do not stop because they are unavailable.
16
17	## Enhanced Advisory Mode
18
19	When the user explicitly enables an API-key-backed reviewer, Codex may request an external advisory pass after the local argument packet exists. Use this only as a second opinion:
20
21	- keep API keys in environment variables, never in CSV, Markdown, logs, or manifests
22	- send only the source subset the user or manifest permits
23	- record the provider, model, source subset, prompt purpose, and output path
24	- place results in advisory notes or reviewer-risk fields
25	- never treat external output as evidence, source support, or a final gate
26
27	## Core Rules
28
29	1. Every contribution must answer a named gap.
30	2. Every main claim must belong to a contribution or the paper-level thesis.
31	3. Every lower-level claim must support a parent claim.
32	4. Every claim must have the right type and amount of evidence.
33	5. Background, methodological, workflow, or governance evidence must not be promoted into empirical, outcome, causal, or deployment validation.
34	6. Important claims must not be evidence-thin; peripheral claims must not be citation-padded.
35	7. Reviewer risks must target a specific intent, gap, contribution, claim, evidence cluster, or limitation.
36	8. Do not fill missing argument links from memory, prior chat, or unstated assumptions. Mark them as gaps.
37	9. Treat external model reviews as advisory notes, not evidence or required gates.
38
39	## Workflow
40
41	### 1. Locate The Manuscript And Evidence Base
42
43	Inspect the manuscript, references, reading notes, evidence registers, figures, tables, and any existing review or release packets. Record which files are canonical and which are exploratory.
44
45	### 2. Build Or Update The Governance Tables
46
47	Create or update these files under `evidence/`:
48
49	- `intent_register.csv`
50	- `contribution_chain.csv`
51	- `claim_hierarchy.csv`
52	- `argument_system_map.csv`
53	- `reviewer_attack_matrix.csv`
54
55	Read `references/argument_schema.md` before creating or editing these files.
56
57	### 3. Check The Argument Hierarchy
58
59	For each paper-level intent:
60
61	1. State the central problem and target reader.
62	2. Name the dominant narrative or missing frame the paper corrects.
63	3. Link each gap to one or more contributions.
64	4. Link each contribution to primary claims.
65	5. Link each claim to evidence IDs, citation keys, or declared evidence gaps.
66	6. Link each high-risk claim to boundary language and reviewer defenses.
67
68	### 4. Audit Evidence Balance
69
70	Use support-balance labels:
71
72	- `unsupported`
73	- `under_supported`
74	- `adequately_supported`
75	- `over_cited`
76	- `misaligned_evidence`
77
78	Treat `over_cited` as a real issue: too many citations on a minor or obvious claim can hide a weak argument spine.
79
80	### 5. Run The Deterministic Checker
81
82	Resolve the bundled helper at `scripts/check_argument_governance.py` relative to this `SKILL.md`, then run it from the project root:
83
84	```bash
85	python3 {skill_dir}/scripts/check_argument_governance.py . --json
86	```
87
88	The checker validates required files, columns, enums, ID references, tree links, missing evidence links, and weak high-risk reviewer defenses. It does not judge scientific novelty or correctness.
89
90	### 6. Report The Argument Health
91
92	For audit-only tasks, produce:
93
94	- gap-contribution alignment issues
95	- claim hierarchy breaks
96	- evidence balance problems
97	- overclaim and boundary-language risks
98	- reviewer attacks with weak defenses
99	- required revision actions
100
101	For edit tasks, wait for user approval before modifying manuscript prose.
102
103	## Output Pattern
104
105	```text
106	## Argument Governance Report
107
108	### Paper Spine
109	- Intent:
110	- Core gap:
111	- Main contributions:
112	- Paper-level thesis:
113
114	### Alignment Issues
115	\| Severity \| Location \| Issue \| Required action \|
116
117	### Claim-Evidence Balance
118	\| Claim ID \| Support balance \| Evidence status \| Risk \| Action \|
119
120	### Reviewer Risks
121	\| Target \| Attack \| Defense strength \| Revision needed \|
122
123	### Next Action

People who liked this also viewed

6 picks

Type

argument-governance

24stars3forks

View on GitHub

How to install argument-governance?

yha9806/academic-writing-toolkit/argument-governance

$npx -y skills add yha9806/academic-writing-toolkit --skill argument-governance

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# /argument-governance - Argument System Governance
2
3	## Purpose
4
5	Make the paper's argument inspectable as a system:
6
7	```text
8	intent -> gap -> contribution -> claim hierarchy -> evidence -> boundary -> reviewer risk
9	```
10
11	Use this before major manuscript revision, before `/peer-review`, before `/self-review`, or after `/manuscript-reframe` when the draft needs stronger gap-contribution and claim-evidence control.
12
13	## Codex-Only Baseline
14
15	Complete this workflow with Codex, local files, and the bundled checker. External reviewers, subagents, Gemini, or other model calls are optional advisory inputs only; do not require them and do not stop because they are unavailable.
16
17	## Enhanced Advisory Mode
18
19	When the user explicitly enables an API-key-backed reviewer, Codex may request an external advisory pass after the local argument packet exists. Use this only as a second opinion:
20
21	- keep API keys in environment variables, never in CSV, Markdown, logs, or manifests
22	- send only the source subset the user or manifest permits
23	- record the provider, model, source subset, prompt purpose, and output path
24	- place results in advisory notes or reviewer-risk fields
25	- never treat external output as evidence, source support, or a final gate
26
27	## Core Rules
28
29	1. Every contribution must answer a named gap.
30	2. Every main claim must belong to a contribution or the paper-level thesis.
31	3. Every lower-level claim must support a parent claim.
32	4. Every claim must have the right type and amount of evidence.
33	5. Background, methodological, workflow, or governance evidence must not be promoted into empirical, outcome, causal, or deployment validation.
34	6. Important claims must not be evidence-thin; peripheral claims must not be citation-padded.
35	7. Reviewer risks must target a specific intent, gap, contribution, claim, evidence cluster, or limitation.
36	8. Do not fill missing argument links from memory, prior chat, or unstated assumptions. Mark them as gaps.
37	9. Treat external model reviews as advisory notes, not evidence or required gates.
38
39	## Workflow
40
41	### 1. Locate The Manuscript And Evidence Base
42
43	Inspect the manuscript, references, reading notes, evidence registers, figures, tables, and any existing review or release packets. Record which files are canonical and which are exploratory.
44
45	### 2. Build Or Update The Governance Tables
46
47	Create or update these files under `evidence/`:
48
49	- `intent_register.csv`
50	- `contribution_chain.csv`
51	- `claim_hierarchy.csv`
52	- `argument_system_map.csv`
53	- `reviewer_attack_matrix.csv`
54
55	Read `references/argument_schema.md` before creating or editing these files.
56
57	### 3. Check The Argument Hierarchy
58
59	For each paper-level intent:
60
61	1. State the central problem and target reader.
62	2. Name the dominant narrative or missing frame the paper corrects.
63	3. Link each gap to one or more contributions.
64	4. Link each contribution to primary claims.
65	5. Link each claim to evidence IDs, citation keys, or declared evidence gaps.
66	6. Link each high-risk claim to boundary language and reviewer defenses.
67
68	### 4. Audit Evidence Balance
69
70	Use support-balance labels:
71
72	- `unsupported`
73	- `under_supported`
74	- `adequately_supported`
75	- `over_cited`
76	- `misaligned_evidence`
77
78	Treat `over_cited` as a real issue: too many citations on a minor or obvious claim can hide a weak argument spine.
79
80	### 5. Run The Deterministic Checker
81
82	Resolve the bundled helper at `scripts/check_argument_governance.py` relative to this `SKILL.md`, then run it from the project root:
83
84	```bash
85	python3 {skill_dir}/scripts/check_argument_governance.py . --json
86	```
87
88	The checker validates required files, columns, enums, ID references, tree links, missing evidence links, and weak high-risk reviewer defenses. It does not judge scientific novelty or correctness.
89
90	### 6. Report The Argument Health
91
92	For audit-only tasks, produce:
93
94	- gap-contribution alignment issues
95	- claim hierarchy breaks
96	- evidence balance problems
97	- overclaim and boundary-language risks
98	- reviewer attacks with weak defenses
99	- required revision actions
100
101	For edit tasks, wait for user approval before modifying manuscript prose.
102
103	## Output Pattern
104
105	```text
106	## Argument Governance Report
107
108	### Paper Spine
109	- Intent:
110	- Core gap:
111	- Main contributions:
112	- Paper-level thesis:
113
114	### Alignment Issues
115	\| Severity \| Location \| Issue \| Required action \|
116
117	### Claim-Evidence Balance
118	\| Claim ID \| Support balance \| Evidence status \| Risk \| Action \|
119
120	### Reviewer Risks
121	\| Target \| Attack \| Defense strength \| Revision needed \|
122
123	### Next Action

People who liked this also viewed

6 picks

Type