0 new in 30d
Momentum is invisible in a flat listing page
Supply-side growth is the quickest way to see which parts of the terminal-agent market are still accumulating inventory.
Benchmarks hub
Read the market before you read the directory.
CmdBrief turns public skill inventory and public release notes into a benchmark surface developers can actually use: category concentration, quality spread, repo language mix, top public listings, and the risk layer that explains what is changing underneath the market.
0tracked skills
0public listings
0new in 30d
0AgentLog entries
Live benchmark signal
0
average quality score across the current public benchmark set
Category leaderLoading...
Market share0%
Breaking releases0
Inventory refreshPending refresh
The benchmark hub combines public footprint, recent supply growth, and release risk instead of pretending a flat registry is enough.
Why benchmarking matters
A directory tells you what exists. A benchmark tells you what matters.
0 average quality
Volume needs a quality read, not just a count
Published coverage only becomes useful when it is paired with quality distribution and public leaderboard context.
0 public listings
Public footprint should be separated from raw coverage
Developers need to know what is public, comparable, and strong enough to earn market attention.
0 breaking releases
Release risk changes what a benchmark means
The hub is useful because AgentLog makes public inventory operational instead of purely descriptive.
Category map
Where public inventory is concentrating right now.
Category benchmark pages will appear here once the statistics endpoint responds.
Release-risk spotlight
The benchmark is more useful when it explains upgrade pressure too.
Breaking changes0
Risky releases0
Safe releases0
Public benchmark pages should help developers evaluate both market shape and operational change, not just browse metadata.
Public footprint
Which listings carry enough public gravity to affect the map?
Leaderboard
Top-starred skills by public footprint
No published leaderboard entries are available yet.
Freshest quality
Newest high-quality skills entering the benchmark set
No recent high-quality skills are available yet.
Inventory health
Read the structure of the market, not just the headline totals.
Quality distribution
How healthy is the public inventory?
Quality buckets will appear once the benchmark API responds.
Average quality score: 0
Repo language mix
Which technical stacks show up most often?
Repo language signal will appear here once the benchmark API responds.
Agent risk coverage
Latest published risk by agent
Agent risk coverage will appear here once AgentLog statistics respond.
Latest AgentLog activity
Recent release notes translated into benchmark context.
Recent AgentLog activity will appear here once the API responds.
Methodology
What this benchmark hub uses right now.
- Skill totals, publication state, and 7d / 30d growth windows.
- Category rollups, category share, and average quality score.
- Quality buckets, repo language mix, and public footprint leaderboards.
- AgentLog counts by agent and risk type, plus the latest published entries.
- Canonical links into category benchmark pages and the skills index.
- Not leading with unstable signals yet: downloads, compatibility, or subscriber leaderboards.
Use the map
Start with the benchmark, then drill into the inventory.
CmdBrief helps developers read public footprint, category momentum, and release risk before they decide which tools deserve deeper evaluation.