CtrlK
BlogDocsLog inGet started
Tessl Logo

postdoc-fellowship-matcher

Match postdoc applicants to eligible fellowships based on nationality and research area

Install with Tessl CLI

npx tessl i github:aipoch/medical-research-skills --skill postdoc-fellowship-matcher
What are skills?

44

Does it follow best practices?

Validation for skill structure

SKILL.md
Review
Evals

Evaluation results

98%

23%

Fellowship Search for a New Postdoc

CLI parameter usage for fellowship matching

Criteria
Without context
With context

Correct script invocation

0%

83%

Nationality parameter

90%

100%

Years post-PhD parameter

90%

100%

Field parameter

90%

100%

Output captured to file

100%

100%

Fellowship names in summary

100%

100%

Deadlines in summary

100%

100%

Requirements included

100%

100%

No reimplementation

0%

100%

No extra packages installed

100%

100%

Without context: $0.5596 · 2m 41s · 24 turns · 94 in / 10,192 out tokens

With context: $0.3485 · 1m 5s · 22 turns · 1,361 in / 3,265 out tokens

97%

21%

Fellowship Matching for a Diverse Lab Group

Multi-applicant matching and fellowship database coverage

Criteria
Without context
With context

Script used for all candidates

0%

100%

Nationality passed per candidate

100%

100%

Years post-PhD per candidate

100%

100%

Field passed per candidate

100%

100%

Institution parameter for US candidate

100%

70%

Fellowship database scope reflected

100%

100%

Deadlines included

100%

100%

Requirements noted

100%

100%

Raw output saved

100%

100%

No reimplementation

0%

100%

Without context: $0.7604 · 3m 22s · 32 turns · 39 in / 12,239 out tokens

With context: $0.6510 · 2m 10s · 34 turns · 1,378 in / 7,583 out tokens

100%

12%

Fellowship Eligibility Check for Late-Career Postdocs

Edge case handling and output format compliance

Criteria
Without context
With context

Script used for both candidates

0%

100%

Correct years for Lefebvre

100%

100%

Correct years for Santos

100%

100%

Nationality passed for both

100%

100%

Field passed for both

100%

100%

No-match case handled

100%

100%

Deadlines in report

100%

100%

Success rate estimates

100%

100%

No extra packages installed

100%

100%

Raw output files saved

100%

100%

Without context: $0.7381 · 3m 13s · 31 turns · 38 in / 12,576 out tokens

With context: $0.3833 · 1m 32s · 22 turns · 1,358 in / 4,759 out tokens

Evaluated
Agent
Claude Code

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.