CtrlK
BlogDocsLog inGet started
Tessl Logo

fmt

Run `make fmt` to check the code format.

Install with Tessl CLI

npx tessl i github:quickwit-oss/quickwit --skill fmt
What are skills?

77

1.69x

Quality

66%

Does it follow best practices?

Impact

95%

1.69x

Average score across 3 eval scenarios

Optimize this skill with Tessl

npx tessl skill review --optimize ./.claude/skills/fmt/SKILL.md
SKILL.md
Review
Evals

Evaluation results

100%

50%

Fix Log Format Violations in Rust Source Files

Log format policy compliance

Criteria
Without context
With context

Uppercase start fixed in indexer.rs

0%

100%

Trailing period fixed in indexer.rs

100%

100%

Uppercase start fixed in searcher.rs

0%

100%

Trailing exclamation fixed in searcher.rs

100%

100%

Compound violation fixed

0%

100%

Correct messages unchanged

0%

100%

Non-log code unchanged

100%

100%

All trailing periods removed

100%

100%

All trailing exclamation marks removed

100%

100%

All uppercase first chars fixed

0%

100%

Without context: $0.4096 · 1m 24s · 18 turns · 25 in / 6,326 out tokens

With context: $0.2469 · 49s · 13 turns · 16 in / 3,792 out tokens

85%

52%

Developer Pre-Commit Format Check Script

Format check workflow script

Criteria
Without context
With context

Uses make fmt command

50%

100%

Runs from quickwit/ subdirectory

0%

100%

Iterative re-run logic

0%

0%

Exit on clean pass

100%

100%

Mentions log format: lowercase

40%

100%

Mentions log format: no trailing punctuation

10%

100%

Correct working directory path

0%

100%

Script is executable / runnable

100%

100%

Without context: $0.2153 · 1m · 14 turns · 20 in / 3,187 out tokens

With context: $0.3179 · 1m 2s · 21 turns · 26 in / 3,314 out tokens

100%

13%

Audit and Fix Log Format Violations Across Rust Modules

Log format audit and fix

Criteria
Without context
With context

All trailing periods removed

100%

100%

All trailing exclamation marks removed

100%

100%

All uppercase-start messages fixed

33%

100%

Correct messages not changed

62%

100%

Report lists violations

90%

100%

Report lists fixes

90%

100%

Trailing comma/semicolon rule not misapplied

100%

100%

All three input files processed

100%

100%

Report covers many violations

100%

100%

Non-log code unchanged

100%

100%

Without context: $0.3891 · 1m 46s · 13 turns · 15 in / 8,656 out tokens

With context: $0.5133 · 1m 45s · 23 turns · 432 in / 7,539 out tokens

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.