claims-extractor

Name: claims-extractor
Author: WILLOSCAR

37installs39forks

How to install claims-extractor?

willoscar/research-units-pipeline-skills/claims-extractor

$npx -y skills add WILLOSCAR/research-units-pipeline-skills --skill claims-extractor

Installs into the current project.

Summary

Use when a review workspace has manuscript text and needs a traceable claim ledger. Trigger: claims extractor, extract claims, contributions, assumptions, peer review, 审稿, 主张提取. Use when: 审稿/评审或 evidence audit，需要把主张列表落盘并可追溯到原文位置（section/page/quote）。 Skip if: 没有可用的稿件/全

Files · 1

View on GitHub

SKILL.md

1	# Claims Extractor
2
3	Transforms manuscript text into a traceable claim ledger for `paper-review`.
4
5	## Input
6
7	- `output/PAPER.md`
8
9	## Output
10
11	- `output/CLAIMS.md`
12	- `output/CLAIMS.jsonl` (`review-claim.v1`, one addressable claim per line)
13
14	## Contract
15
16	Each claim block must include:
17	- claim text
18	- type: `empirical` or `conceptual`
19	- scope
20	- source pointer back into `output/PAPER.md`
21
22	The Markdown and JSONL files are two views of the same claim records. Downstream
23	skills should prefer JSONL and retain Markdown fallback for older Workspaces.
24
25	## Script boundary
26
27	`scripts/run.py` should:
28	- detect claim-like sentences
29	- normalize them into stable claim blocks
30	- attach source pointers
31	- separate empirical and conceptual claims
32
33	Keep parsing and ranking heuristics in shared review tooling, not in the skill script.
34
35	## Acceptance
36
37	- `output/CLAIMS.md` exists
38	- `output/CLAIMS.jsonl` exists and every record has a stable `claim_id`
39	- every claim has a source pointer
40	- empirical and conceptual claims are separated
41
42	## Non-goals
43
44	- judging whether a claim is good
45	- generating evidence gaps
46	- writing review prose

People who liked this also viewed

6 picks

Type

claims-extractor

37installs39forks

View on GitHub

How to install claims-extractor?

willoscar/research-units-pipeline-skills/claims-extractor

$npx -y skills add WILLOSCAR/research-units-pipeline-skills --skill claims-extractor

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Claims Extractor
2
3	Transforms manuscript text into a traceable claim ledger for `paper-review`.
4
5	## Input
6
7	- `output/PAPER.md`
8
9	## Output
10
11	- `output/CLAIMS.md`
12	- `output/CLAIMS.jsonl` (`review-claim.v1`, one addressable claim per line)
13
14	## Contract
15
16	Each claim block must include:
17	- claim text
18	- type: `empirical` or `conceptual`
19	- scope
20	- source pointer back into `output/PAPER.md`
21
22	The Markdown and JSONL files are two views of the same claim records. Downstream
23	skills should prefer JSONL and retain Markdown fallback for older Workspaces.
24
25	## Script boundary
26
27	`scripts/run.py` should:
28	- detect claim-like sentences
29	- normalize them into stable claim blocks
30	- attach source pointers
31	- separate empirical and conceptual claims
32
33	Keep parsing and ranking heuristics in shared review tooling, not in the skill script.
34
35	## Acceptance
36
37	- `output/CLAIMS.md` exists
38	- `output/CLAIMS.jsonl` exists and every record has a stable `claim_id`
39	- every claim has a source pointer
40	- empirical and conceptual claims are separated
41
42	## Non-goals
43
44	- judging whether a claim is good
45	- generating evidence gaps
46	- writing review prose

People who liked this also viewed

6 picks

Type