Collect and normalize agent logs, discover installed verifiers, and dispatch LLM judges to evaluate adherence. Produces per-session verdicts and aggregated reports.
91
90%
Does it follow best practices?
Impact
96%
3.09xAverage score across 3 eval scenarios
Passed
No known issues
Companion tile outside .tessl/
0%
100%
tile.json present
0%
100%
docs key in tile.json
0%
100%
docs/overview.md present
0%
37%
Every verifier linked in overview
0%
100%
Flat verifiers dir at tile root
0%
100%
Sources field included
0%
100%
Kebab-case file names
50%
100%
watch-local install command
0%
0%
Binary checklist rules
0%
100%
SCRIPTS_PATH find pattern
0%
100%
Correct script path in find
0%
100%
uv run python3 invocation
0%
100%
SCRIPTS_PATH empty check
0%
100%
max-sessions flag
0%
100%
search-before-audit pattern
33%
100%
Small batch for search results
50%
100%
No auto Phase 2 progression
100%
100%
Verifiers in skill dir
100%
100%
Flat verifiers dir
100%
100%
Kebab-case file names
100%
100%
Sources field omitted
100%
100%
Activation verifier
100%
100%
Binary checklist rules
50%
100%
Multi-part decomposition
20%
100%
Specific decision-point relevant_when
0%
100%
Context fields present
0%
100%
Prohibition verifier present
100%
100%