sensei
**WORKFLOW SKILL** — Iteratively improve skill frontmatter compliance using the Ralph loop pattern. WHEN: "run sensei", "sensei help", "improve skill", "fix frontmatter", "skill compliance", "frontmatter audit", "score skill", "check skill tokens". INVOKES: token counting tools, test runners, git commands. FOR SINGLE OPERATIONS: use token CLI directly for counts/checks.
Install with Tessl CLI
npx tessl i github:microsoft/github-copilot-for-azure --skill sensei95
Does it follow best practices?
Evaluation — 100%
↑ 1.63xAgent success when using this skill
Validation for skill structure
Evaluation results
100%
56%Azure Skill Frontmatter Modernization
Frontmatter compliance improvement
Inline double-quoted string
0%
100%
WHEN: trigger format
0%
100%
Quoted trigger phrases
70%
100%
No DO NOT USE FOR
100%
100%
Word count constraint
0%
100%
Description length minimum
100%
100%
Action verb lead
70%
100%
License recommendation
0%
100%
Version recommendation
100%
100%
Without context: $0.5339 · 1m 49s · 25 turns · 25 in / 5,721 out tokens
With context: $0.5555 · 1m 27s · 19 turns · 5,711 in / 4,566 out tokens
100%
20%Adding Trigger Test Coverage for Azure Key Vault Skill
Test scaffolding and trigger coverage
Correct file path
100%
100%
SKILL_NAME constant
100%
100%
Trigger prompt count
100%
100%
Trigger prompt alignment
100%
100%
Anti-trigger prompt count
100%
100%
Unrelated topic anti-triggers
100%
100%
Other cloud provider anti-triggers
0%
100%
Related Azure service anti-triggers
0%
100%
Valid TypeScript structure
100%
100%
Without context: $0.1540 · 31s · 10 turns · 15 in / 1,836 out tokens
With context: $0.3200 · 53s · 15 turns · 54 in / 2,673 out tokens
100%
41%Azure Skills Frontmatter Compliance Audit
Compliance scoring and workflow commands
Name validation failure
100%
100%
Low score for no triggers
75%
100%
Medium-High score for WHEN
100%
100%
Flags DO NOT USE FOR risk
75%
100%
Flags folded scalar
0%
100%
License recommendation
100%
100%
Version recommendation
100%
100%
Correct test command
0%
100%
Correct token check command
0%
100%
No DO NOT USE FOR in recommendations
100%
100%
Correct references validation command
0%
100%
Without context: $0.4064 · 2m 14s · 17 turns · 23 in / 7,704 out tokens
With context: $0.6161 · 2m 5s · 16 turns · 7,835 in / 7,534 out tokens
- Evaluated
- Agent
- Claude Code
- Model
- Unknown
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.