Performance review instructions for the expert agent
66
Quality
50%
Does it follow best practices?
Impact
93%
2.81xAverage score across 3 eval scenarios
Optimize this skill with Tessl
npx tessl skill review --optimize ./dot_config/opencode/skill/review-performance/SKILL.mdExploration log and N+1 detection
Exploration log present
0%
100%
Full file read documented
0%
100%
Schema check documented
0%
100%
Association chain traced
0%
100%
N+1 for customer reported
70%
100%
N+1 for line_items reported
70%
100%
JSON object output
100%
100%
Both required top-level keys
0%
100%
Finding fields complete
25%
100%
Valid severity values
0%
100%
I-statement body
0%
100%
2-5 word titles
0%
100%
Without context: $0.2402 · 1m 26s · 8 turns · 13 in / 4,733 out tokens
With context: $0.6329 · 3m 7s · 23 turns · 275 in / 10,677 out tokens
Pre-existing issues and scope enforcement
Missing index on name columns
100%
100%
Pre-existing severity for role
0%
100%
Role index finding present
100%
100%
Security issue escalated
0%
100%
Security not in findings
0%
100%
Escalation for_reviewer field
0%
100%
Escalation file and line
0%
100%
Schema check in log
62%
100%
Git blame check in log
62%
100%
Pagination not reported
0%
0%
2-5 word titles
0%
100%
Without context: $0.2388 · 1m 40s · 10 turns · 15 in / 4,877 out tokens
With context: $0.3761 · 1m 52s · 12 turns · 260 in / 6,637 out tokens
Algorithmic complexity and pagination
O(n²) complexity reported
85%
85%
Unbounded loop identified
87%
100%
N+1 inside compute_similarity
100%
100%
Pagination finding
90%
100%
Loop analysis in exploration log
0%
100%
I-statement bodies
0%
100%
Question framing
0%
0%
2-5 word titles
0%
70%
No style or correctness findings
100%
100%
Test file checked
0%
100%
Both required top-level keys
0%
100%
Without context: $0.3136 · 1m 44s · 10 turns · 15 in / 5,171 out tokens
With context: $0.3529 · 1m 57s · 12 turns · 172 in / 5,843 out tokens
Table of Contents
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.