writing-skills
Use when creating new skills, editing existing skills, or verifying skills work before deployment
71
Quality
58%
Does it follow best practices?
Impact
94%
1.77xAverage score across 3 eval scenarios
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/writing-skills/SKILL.mdEvaluation results
86%
82%New Skill: Async Database Connection Pooling
SKILL.md frontmatter and CSO-optimized description
Frontmatter fields only
0%
100%
Name format
0%
100%
Verb-first name
0%
0%
Description starts with Use when
0%
100%
Description is third person
0%
100%
Description has no workflow summary
0%
100%
Description focuses on triggering conditions
0%
100%
Description under 500 characters
0%
100%
Frontmatter under 1024 chars
0%
100%
Single code example
0%
40%
Notes justify frontmatter decisions
40%
100%
Without context: $0.3428 · 1m 42s · 16 turns · 18 in / 5,086 out tokens
With context: $0.4983 · 1m 44s · 20 turns · 27 in / 5,223 out tokens
100%
18%Structuring the pptx-generation Skill
Skill file organization, token efficiency, cross-referencing, and flowchart usage
Flat namespace directory
100%
100%
API reference in separate file
100%
100%
Template script in separate file
100%
100%
SKILL.md stays concise
100%
100%
CLI flags use --help reference
60%
100%
Cross-reference without @ link
100%
100%
Cross-reference has explicit requirement marker
25%
100%
Single-language code example
100%
100%
No flowchart for linear workflow
100%
100%
Frontmatter cleaned up
0%
100%
Rationale explains decisions
100%
100%
Without context: $0.5919 · 3m 5s · 21 turns · 28 in / 11,436 out tokens
With context: $0.7100 · 2m 52s · 23 turns · 26 in / 10,197 out tokens
96%
23%Deploying a Code Review Enforcement Skill
Iron Law TDD cycle and bulletproofing discipline-enforcing skills
Rationalization table present
80%
100%
Rationalizations from baseline
100%
100%
Explicit negations for loopholes
100%
100%
Red flags list
50%
100%
Spirit vs letter addressed
37%
100%
Authority language used
100%
100%
No liking/reciprocity language
100%
100%
Pressure scenarios have multi-pressure design
100%
100%
Pressure scenarios force explicit choice
0%
100%
Deployment checklist uses todo format
100%
100%
Description has violation symptoms
62%
100%
Commitment mechanism present
33%
33%
Without context: $0.4073 · 2m 17s · 21 turns · 27 in / 6,780 out tokens
With context: $0.7130 · 2m 37s · 24 turns · 28 in / 7,637 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.