Observed arrival · 2026-11-01
Knackdex shows agent-skill outputs before installation
An index of 119 agent skills pairs entries with sample outputs labeled by model and version, letting readers inspect results before installing a skill.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- People choosing or building agent skills
- Worth noticing
- The unslop example includes a task prompt, model and version labels, run details, and a checklist of claims to verify.
Field notes
The featured “#111 unslop” record includes the task prompt, generated revision, model labels, and run context, followed by checks for claims that need substantiation. One visible Claude Code run reports 55,644 input tokens and 2,715 output tokens; its note clarifies that the input count includes the tool’s system prompt, a useful caveat when interpreting the run.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
One card from the complete issue