CtrlK
BlogDocsLog inGet started
Tessl Logo

conflict-of-interest-checker

Check for co-authorship conflicts between authors and suggested reviewers

68

1.94x
Quality

52%

Does it follow best practices?

Impact

99%

1.94x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./scientific-skills/Academic Writing/conflict-of-interest-checker/SKILL.md
SKILL.md
Quality
Evals
Security

Evaluation results

98%

62%

Reviewer Conflict Screen for Cardiology Manuscript

Script invocation with correct flags

Criteria
Without context
With context

Correct script path

0%

83%

--authors flag used

0%

100%

--reviewers flag used

0%

100%

Comma-separated names

0%

100%

All three authors included

50%

100%

All five reviewers included

50%

100%

No extra packages installed

100%

100%

Script not reimplemented

0%

100%

Recommendations captured

100%

100%

Report output present

100%

100%

Without context: $0.3953 · 1m 59s · 25 turns · 124 in / 6,319 out tokens

With context: $0.2585 · 50s · 16 turns · 1,995 in / 2,481 out tokens

100%

53%

Co-authorship Conflict Check Using Publication History

CSV format and --publications flag

Criteria
Without context
With context

CSV column: author

100%

100%

CSV column: reviewer

0%

100%

CSV column: paper_id

100%

100%

--publications flag used

0%

100%

Correct script invocation

0%

100%

Script not reimplemented

0%

100%

Ferreira-Tanaka conflict captured

50%

100%

Zhang-Popescu conflict captured

50%

100%

Conflicts flagged in output

75%

100%

No extra packages installed

100%

100%

Without context: $0.4831 · 1m 55s · 26 turns · 28 in / 7,319 out tokens

With context: $0.5930 · 2m 18s · 24 turns · 2,003 in / 8,421 out tokens

99%

29%

Peer Review Integrity Audit for Annual Conference

Output interpretation and recommendations

Criteria
Without context
With context

Correct CSV columns

0%

100%

--publications flag used

0%

100%

Correct script path

0%

87%

Three separate runs

100%

100%

O'Brien flagged (Paper 1)

100%

100%

Nakamura flagged (Paper 2)

100%

100%

Recuse recommendation documented

100%

100%

Shared papers cited

100%

100%

Comma-separated names format

100%

100%

No extra packages

100%

100%

Compliance use case acknowledged

100%

100%

Without context: $0.5878 · 2m 24s · 28 turns · 37 in / 9,243 out tokens

With context: $0.6184 · 2m 10s · 27 turns · 2,005 in / 8,248 out tokens

Repository
aipoch/medical-research-skills
Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.