PulseGatePost-LLM software, agents & workflows market (since 2022)
Coverage
166,372

Apps indexed

 

Freshness
48 min ago

Last update

 

Cadence
1,427/day

7-day average

Indexed today: 480

PulseGate

Market catalog for public software products, models, infra, and workflow tools.

Software is shipping faster than ever, and a growing share of it lives outside the official app stores. PulseGate is a free public catalog — built for builders, analysts, and everyday users.

Platform

  • All Apps
  • Categories
  • Industry Updates
  • Data Sources
  • Coverage Rules
  • Glossary
  • Embed Widget

Support

  • Help Center
  • Submit a Project
  • Report an Issue

Company

  • About
  • Press & Data
  • Contact
  • Platform Status

Legal

  • Privacy
  • Terms
  • Disclaimer

© 2026 PulseGate. Operated by Dymaxio s.r.o., Prague, Czech Republic.·

All systems operational
PulseGate
SK
skill-eval-harness
Visit ↗
  1. Home/
  2. LLM eval & observability/
  3. skill-eval-harness
←Back to results
S

skill-eval-harness

pypi.org·Infrastructure

skill-eval-harness is an open-source CLI tool for evaluating agent skills using benchmarks, paired variants, and holdout splits. It supports repeated-run statistics, script assertions, Anthropic-compatible exports, and Jetty adapter integration, targeting AI researchers and developers.

Open SourceMIT
CLI
S
Visit PyPI↗
⭐54
stars
🍴3
forks
✓7
features
📅2026
since

Overview

7 features
  • ✓Skill evaluation
  • ✓Paired variants
  • ✓Holdout splits
  • ✓Repeated-run stats
  • ✓Script assertions
  • ✓Anthropic export
  • ✓Jetty adapter

Tags

agent-evaluationbenchmarkinganthropic-exportjetty-adapterskill-testing

AI capabilities

CodeStructured

Built with & integrations

Runs on
CLI

Trust & compliance

LicenseMIT
Verified signals
✓ HTTPS✓ Open Source✓ Free tier✓ GitHub · ★ 54✓ Active maintenance

Recent events

Latest indexed changes and source events

  1. IndexedJul 8, 10:34 AM

    skill-eval-harness discovered by the PulseGate indexer

    Source: PulseGate indexerOpen ↗

At a glance

Platforms
Cli
Languages
English
Open source
Yes · ★ 54
License
MIT
First seen
Jul 8, 2026
Activity
🟢 Active
Status
🟢 Active
Built for
AI researchers and developers evaluating agent performance
Model
Open source
Solves
Standardizing and automating the evaluation of agent skills across benchmarks and variants.

Developer

adewale
Small team
↗ GitHub

Open source

View on GitHub →
⭐ Stars
54
🍴 Forks
3
Open issues
0
Last commit
3d ago
Commits 90d
76
Contributors
2
Authorship
Small team
Default branch
main
Latest release
v0.5.1 · 3d ago

PulseGate index

Confidence
Low · 64
Indexed
Jul 8, 2026
Lifecycle
Alive
Activity
Active
First seen
Jul 2026
Last seen
3d ago
Freshness
Unknown
Identity audit (9)
Entity ID
cmrbxyi6100kjjqxtsk84tesl
Slug
skill-eval-harness-pypi-org
Verification state
Indexed for public listing
Claim / listing state
Unclaimed · listed: yes
Index status
Included in index
Latest evidence snapshot
Jul 8, 2026
Timeline basis
Indexed-at chronology (no inferred launch/funding milestones).
Last updated
Jul 8, 2026
Canonical URL
https://pypi.org/project/skill-eval-harness

Similar apps

Other apps tracked under the same category.

  • skillfed
    pypi.org
  • agent-skill-eval
    pypi.org
  • skill-auditor
    pypi.org
  • skill-lab
    github.com
  • skillstat
    pypi.org
  • skill-deployer
    pypi.org

See 9 alternatives to skill-eval-harness