CtrlK
BlogDocsLog inGet started
Tessl Logo

rationalize-deps

Analyze Cargo.toml dependencies and attempt to remove unused features to reduce compile times and binary size

Install with Tessl CLI

npx tessl i github:quickwit-oss/tantivy --skill rationalize-deps
What are skills?

74

1.40x

Quality

72%

Does it follow best practices?

Impact

62%

1.40x

Average score across 3 eval scenarios

Optimize this skill with Tessl

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

Evaluation results

73%

23%

Reduce Compile Times by Rationalizing Rust Dependencies

Dependency feature rationalization workflow

Criteria
Without context
With context

serde default-features disabled

0%

100%

tokio default-features disabled

0%

100%

serde derive feature kept

100%

100%

tokio macros feature kept

100%

100%

tokio rt feature kept

100%

100%

cargo check command documented

0%

0%

cargo tree command used

0%

0%

cargo metadata command used

0%

0%

original config documented

100%

100%

new config documented

100%

100%

features removed documented

25%

62%

required features documented

100%

100%

Without context: $0.2169 · 1m 6s · 9 turns · 14 in / 3,811 out tokens

With context: $0.6746 · 2m 38s · 29 turns · 192 in / 9,695 out tokens

57%

2%

Minimize Reqwest Features for a Lightweight HTTP Client

Binary search for minimal feature set

Criteria
Without context
With context

reqwest default-features disabled

100%

100%

cargo metadata used

0%

0%

binary search methodology

0%

0%

iterative compilation checks

0%

0%

default features listed

75%

75%

rustls-tls or tls feature selected

100%

100%

json feature selected

100%

100%

unnecessary features removed

100%

100%

feature justifications provided

87%

100%

final verification commands

0%

25%

features removed listed

80%

80%

Without context: $0.3214 · 1m 25s · 16 turns · 21 in / 4,838 out tokens

With context: $0.7531 · 3m 10s · 37 turns · 294 in / 10,352 out tokens

56%

28%

Audit and Reduce a Rust Workspace's Dependency Footprint

Workspace-wide analysis, prioritization, and full build verification

Criteria
Without context
With context

serde rationalized in core

0%

100%

tokio rationalized in core

100%

100%

tokio rationalized in cli

42%

100%

clap rationalized

33%

100%

serde rationalized in utils

0%

100%

cargo tree -d for duplicates

0%

22%

prioritization rationale

25%

37%

dev-dependencies consideration

77%

100%

final cargo check --all-targets

0%

0%

cargo test --no-run

0%

0%

per-dependency docs

88%

100%

cargo tree features command

0%

0%

cargo metadata command

0%

0%

Without context: $0.4867 · 2m 1s · 21 turns · 28 in / 7,436 out tokens

With context: $0.7912 · 3m 36s · 34 turns · 31 in / 13,996 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.