citecheck

Name: citecheck
Author: color4-alt

53stars1forks

How to install citecheck?

color4-alt/citecheck/citecheck

$npx -y skills add color4-alt/CiteCheck --skill citecheck

Installs into the current project.

Summary

Use when the user asks to "verify citations", "check references", "validate paper citations", or "evaluate reference relevance". It extracts references from LaTeX/PDF papers, checks formatting rules, verifies existence via Crossref / Semantic Scholar / OpenAlex / PubMed / arXiv /

Files · 1

View on GitHub

SKILL.md

1	# CiteCheck — Paper Citation Verification
2
3	## Overview
4
5	CiteCheck verifies academic paper citations by combining structured parsing with agent-native LLM evaluation. It supports LaTeX source files (preferred) and PDF fallback.
6
7	## Workflow
8
9	1. Parse paper: Call `citecheck` CLI to read LaTeX (preferred) or PDF, extract references and body text
10	2. Format check: Call `citecheck` CLI to validate bibliography entries
11	3. Queryability verification: Call `citecheck` CLI to verify existence via Crossref / Semantic Scholar
12	4. Evaluate thematic relevance: Evaluate directly using the agent's reasoning capabilities — compare cited paper title/abstract/venue against the citing paper
13	5. Evaluate semantic accuracy: Evaluate directly using the agent's reasoning capabilities — compare in-text citation context against cited source content
14	6. Generate report: Aggregate all results into a Markdown report
15
16	> Why matching steps are not done by the CLI
17	> The `citecheck` CLI can run standalone with optional `--api-key` for external LLM-powered matching. When used as a Skill, the host agent itself possesses LLM reasoning capabilities. Direct evaluation is faster, more consistent, and requires no additional API keys from the user.
18
19	---
20
21	## 1. Parse Paper (CLI)
22
23	### LaTeX source (preferred)
24
25	```bash
26	citecheck path/to/latex_project/ --skip-verification --skip-semantic -o parsed_report.md
27	```
28
29	Or parse a single file:
30
31	```bash
32	citecheck main.tex --skip-verification --skip-semantic -o parsed_report.md
33	```
34
35	### PDF (fallback)
36
37	```bash
38	citecheck paper.pdf --skip-verification --skip-semantic -o parsed_report.md
39	```
40
41	> `--skip-verification` and `--skip-semantic` are required in Skill mode because steps 3–5 are performed directly by the agent.
42
43	---
44
45	## 2. Format Check (CLI)
46
47	Call `citecheck` to check and report format issues:
48
49	\| Check item \| Description \|
50	\|------------\|-------------\|
51	\| Required fields \| Author, title, year, venue completeness \|
52	\| Format consistency \| Punctuation, capitalization, abbreviation uniformity \|
53	\| DOI/URL \| If present, whether format is correct and accessible \|
54	\| Year sanity \| No `202x` placeholders, not in the future \|
55
56	Detailed rules: see [./references/format-check-rules.md](./references/format-check-rules.md).
57
58	---
59
60	## 3. Queryability Verification (CLI + Agent Supplement)
61
62	Call `citecheck` to query Crossref and Semantic Scholar public APIs for citation existence.
63
64	API details: see [./references/api-reference.md](./references/api-reference.md).
65
66	If the CLI fails due to network/SSL issues, use WebSearch to directly query suspicious citations (especially those marked "unverifiable"), supplementing the verification results.
67
68	---
69
70	## 4. Evaluate Thematic Relevance (Agent Direct)
71
72	Do not call the CLI or any external API. Evaluate directly using the agent's reasoning capabilities.
73
74	For each reference, extract:
75	- Citing paper: `title`, `abstract`, `keywords`
76	- Cited paper: `title`, `abstract` (from API results or WebSearch), `venue`
77
78	Evaluation prompt template: see [./references/thematic-scoring-prompt.md](./references/thematic-scoring-prompt.md).
79
80	---
81
82	## 5. Evaluate Semantic Accuracy (Agent Direct)
83
84	Do not call the CLI or any external API. Evaluate directly using the agent's reasoning capabilities.
85
86	For each in-text citation position:
87
88	1. Extract 1–2 sentences before and after the citation marker as `citing_text`
89	2. Obtain the cited paper's abstract via Semantic Scholar or WebSearch as `cited_text`
90	3. Evaluate semantic consistency using the prompt template: see [./references/semantic-matching-prompt.md](./references/semantic-matching-prompt.md).
91
92	---
93
94	## 6. Output Report
95
96	Generate a Markdown report containing:
97
98	### Summary
99	- Total references
100	- Format issues count
101	- Query failures / suspicious count
102	- Average thematic relevance score
103	- Average semantic accuracy score
104
105	### Detailed Results Table
106
107	\| No. \| Title \| Format \| Queryable \| Thematic \| Semantic \| Notes \|
108	\|-----\|-------\|--------\|-----------\|----------\|----------\|-------\|
109	\| 1 \| ... \| ✅/⚠️/❌ \| ✅/❌ \| 0.85 \| 0.90 \| DOI mismatch \|
110
111	### Issue Summary
112	- List all findings and recommendations by severity
113
114	---
115
116	## Dependencies
117
118	```bash
119	pip install CiteCheck
120	```
121
122	For PDF support:
123
124	```bash
125	pip install CiteCheck[pdf]
126	```
127
128	---
129
130	## Notes
131
132	- No `--api-key` is required in Skill mode. Thematic and semantic matching are performed directly by the agent, with no external LLM API calls.
133	- Prioritize LaTeX source files; parsing accuracy is far higher than PDF.
134	- Add delays and retries to API calls to avoid rate limiting.
135	- When Semantic Sch

People who liked this also viewed

6 picks

Type

citecheck

53stars1forks

View on GitHub

How to install citecheck?

color4-alt/citecheck/citecheck

$npx -y skills add color4-alt/CiteCheck --skill citecheck

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# CiteCheck — Paper Citation Verification
2
3	## Overview
4
5	CiteCheck verifies academic paper citations by combining structured parsing with agent-native LLM evaluation. It supports LaTeX source files (preferred) and PDF fallback.
6
7	## Workflow
8
9	1. Parse paper: Call `citecheck` CLI to read LaTeX (preferred) or PDF, extract references and body text
10	2. Format check: Call `citecheck` CLI to validate bibliography entries
11	3. Queryability verification: Call `citecheck` CLI to verify existence via Crossref / Semantic Scholar
12	4. Evaluate thematic relevance: Evaluate directly using the agent's reasoning capabilities — compare cited paper title/abstract/venue against the citing paper
13	5. Evaluate semantic accuracy: Evaluate directly using the agent's reasoning capabilities — compare in-text citation context against cited source content
14	6. Generate report: Aggregate all results into a Markdown report
15
16	> Why matching steps are not done by the CLI
17	> The `citecheck` CLI can run standalone with optional `--api-key` for external LLM-powered matching. When used as a Skill, the host agent itself possesses LLM reasoning capabilities. Direct evaluation is faster, more consistent, and requires no additional API keys from the user.
18
19	---
20
21	## 1. Parse Paper (CLI)
22
23	### LaTeX source (preferred)
24
25	```bash
26	citecheck path/to/latex_project/ --skip-verification --skip-semantic -o parsed_report.md
27	```
28
29	Or parse a single file:
30
31	```bash
32	citecheck main.tex --skip-verification --skip-semantic -o parsed_report.md
33	```
34
35	### PDF (fallback)
36
37	```bash
38	citecheck paper.pdf --skip-verification --skip-semantic -o parsed_report.md
39	```
40
41	> `--skip-verification` and `--skip-semantic` are required in Skill mode because steps 3–5 are performed directly by the agent.
42
43	---
44
45	## 2. Format Check (CLI)
46
47	Call `citecheck` to check and report format issues:
48
49	\| Check item \| Description \|
50	\|------------\|-------------\|
51	\| Required fields \| Author, title, year, venue completeness \|
52	\| Format consistency \| Punctuation, capitalization, abbreviation uniformity \|
53	\| DOI/URL \| If present, whether format is correct and accessible \|
54	\| Year sanity \| No `202x` placeholders, not in the future \|
55
56	Detailed rules: see [./references/format-check-rules.md](./references/format-check-rules.md).
57
58	---
59
60	## 3. Queryability Verification (CLI + Agent Supplement)
61
62	Call `citecheck` to query Crossref and Semantic Scholar public APIs for citation existence.
63
64	API details: see [./references/api-reference.md](./references/api-reference.md).
65
66	If the CLI fails due to network/SSL issues, use WebSearch to directly query suspicious citations (especially those marked "unverifiable"), supplementing the verification results.
67
68	---
69
70	## 4. Evaluate Thematic Relevance (Agent Direct)
71
72	Do not call the CLI or any external API. Evaluate directly using the agent's reasoning capabilities.
73
74	For each reference, extract:
75	- Citing paper: `title`, `abstract`, `keywords`
76	- Cited paper: `title`, `abstract` (from API results or WebSearch), `venue`
77
78	Evaluation prompt template: see [./references/thematic-scoring-prompt.md](./references/thematic-scoring-prompt.md).
79
80	---
81
82	## 5. Evaluate Semantic Accuracy (Agent Direct)
83
84	Do not call the CLI or any external API. Evaluate directly using the agent's reasoning capabilities.
85
86	For each in-text citation position:
87
88	1. Extract 1–2 sentences before and after the citation marker as `citing_text`
89	2. Obtain the cited paper's abstract via Semantic Scholar or WebSearch as `cited_text`
90	3. Evaluate semantic consistency using the prompt template: see [./references/semantic-matching-prompt.md](./references/semantic-matching-prompt.md).
91
92	---
93
94	## 6. Output Report
95
96	Generate a Markdown report containing:
97
98	### Summary
99	- Total references
100	- Format issues count
101	- Query failures / suspicious count
102	- Average thematic relevance score
103	- Average semantic accuracy score
104
105	### Detailed Results Table
106
107	\| No. \| Title \| Format \| Queryable \| Thematic \| Semantic \| Notes \|
108	\|-----\|-------\|--------\|-----------\|----------\|----------\|-------\|
109	\| 1 \| ... \| ✅/⚠️/❌ \| ✅/❌ \| 0.85 \| 0.90 \| DOI mismatch \|
110
111	### Issue Summary
112	- List all findings and recommendations by severity
113
114	---
115
116	## Dependencies
117
118	```bash
119	pip install CiteCheck
120	```
121
122	For PDF support:
123
124	```bash
125	pip install CiteCheck[pdf]
126	```
127
128	---
129
130	## Notes
131
132	- No `--api-key` is required in Skill mode. Thematic and semantic matching are performed directly by the agent, with no external LLM API calls.
133	- Prioritize LaTeX source files; parsing accuracy is far higher than PDF.
134	- Add delays and retries to API calls to avoid rate limiting.
135	- When Semantic Sch

People who liked this also viewed

6 picks

Type