CtrlK
BlogDocsLog inGet started
Tessl Logo

screenshot

Find and display recent screenshots. Triggers: screenshot, check screenshot, show screenshot, recent screenshot, last screenshot.

Install with Tessl CLI

npx tessl i github:NeverSight/skills_feed --skill screenshot
What are skills?

85

Does it follow best practices?

Validation for skill structure

SKILL.md
Review
Evals

Evaluation results

100%

9%

Screenshot Finder Utility

Platform-specific directory detection order

Criteria
Without context
With context

Linux path priority 1

100%

100%

Linux path priority 2

100%

100%

Linux path priority 3

100%

100%

macOS Desktop first

100%

100%

First-match selection

100%

100%

PNG support

100%

100%

Multiple format support

100%

100%

Top-level only

100%

100%

Modification-time sort

100%

100%

No-directory message

100%

100%

Checked-locations list

0%

100%

No-images message

100%

100%

Output file produced

100%

100%

Without context: $0.3480 · 1m 26s · 22 turns · 29 in / 4,855 out tokens

With context: $0.4650 · 1m 40s · 26 turns · 64 in / 5,257 out tokens

100%

36%

Screenshot Catalog Reporter

Structured output format and metadata display

Criteria
Without context
With context

Directory header

0%

100%

Numbered sections

70%

100%

File metadata field

100%

100%

Size metadata field

100%

100%

Modified metadata field

0%

100%

Default count of 5

50%

100%

Newest first ordering

0%

100%

Image-only filtering

100%

100%

Limited count report

100%

100%

webp and jpeg support

100%

100%

No subdirectory recursion

100%

100%

Without context: $0.3299 · 1m 10s · 21 turns · 28 in / 4,167 out tokens

With context: $0.4982 · 1m 47s · 22 turns · 24 in / 7,113 out tokens

95%

20%

Screenshot Tool Edge Case Handler

Edge case messages and custom directory config

Criteria
Without context
With context

No-directory message

100%

100%

Checked locations listed

0%

50%

No-images message

100%

100%

No-images extensions listed

0%

100%

Count-exceeds message

100%

100%

All available shown

100%

100%

Config file created

100%

100%

Config default_count field

100%

100%

Config file_extensions field

0%

100%

Config read at runtime

100%

100%

Test harness runs all cases

100%

100%

Without context: $0.3911 · 1m 30s · 24 turns · 30 in / 5,476 out tokens

With context: $0.6652 · 2m 20s · 30 turns · 116 in / 9,203 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.