Optimize your skills and tiles: review SKILL.md quality, generate eval scenarios, run evals, compare across models, diagnose gaps, and re-run until scores improve.
88
94%
Does it follow best practices?
Impact
88%
1.07xAverage score across 24 eval scenarios
Passed
No known issues
{
"context": "Tests whether the agent correctly identifies context files for tessl scenario generation — including .mdc rule files, agent guidance files, and tile config — while excluding cached tile content in .tessl/, generic documentation, and source code.",
"type": "weighted_checklist",
"checklist": [
{
"name": "identifies_mdc_files",
"description": "Includes the three .mdc files (.cursor/rules/api-patterns.mdc, .cursor/rules/testing.mdc, .cursor/rules/component-style.mdc) as context files, recognizing them as agent coding rules",
"max_score": 15
},
{
"name": "identifies_claude_md",
"description": "Includes CLAUDE.md as a context file, recognizing it as agent-specific project guidance",
"max_score": 10
},
{
"name": "identifies_agents_md",
"description": "Includes docs/AGENTS.md as a context file, recognizing it as agent-specific project guidance",
"max_score": 10
},
{
"name": "identifies_tessl_json",
"description": "Includes tessl.json as a context file, recognizing it as tile configuration",
"max_score": 10
},
{
"name": "excludes_tessl_cache",
"description": "Does NOT include any files under .tessl/tiles/ (e.g., .tessl/tiles/web-app-helper/SKILL.md) as context. The .tessl/ directory is a local cache of installed tiles, not a project context file",
"max_score": 15
},
{
"name": "excludes_generic_docs",
"description": "Does NOT include README.md or CHANGELOG.md as context files — these are general project docs, not agent guidance or coding conventions",
"max_score": 10
},
{
"name": "excludes_source_and_build_config",
"description": "Does NOT include source files (src/utils/helpers.ts, src/components/Button.tsx, etc.) or build config (tsconfig.json, package.json) as context files",
"max_score": 10
},
{
"name": "constructs_valid_context_flag",
"description": "Produces a syntactically valid tessl scenario generate command with a --context flag that would match the identified context files. The pattern should cover *.mdc, CLAUDE.md, AGENTS.md, and tessl.json — either as a glob pattern (e.g., --context=\"*.mdc,CLAUDE.md,docs/AGENTS.md,tessl.json\") or as individual entries. The command must also include --commits=a1b2c3d,e4f5g6h and --workspace main-ws",
"max_score": 20
}
]
}evals
scenario-1
scenario-2
scenario-3
scenario-4
scenario-5
scenario-6
scenario-7
scenario-8
scenario-9
scenario-10
scenario-11
scenario-12
scenario-13
scenario-14
scenario-15
scenario-16
scenario-17
scenario-18
scenario-19
scenario-20
scenario-21
scenario-22
scenario-23
scenario-24
skills
compare-skill-model-performance
optimize-skill-instructions
references
optimize-skill-performance
optimize-skill-performance-and-instructions