Skip to main content
agentic-evalby GitHubDevelopmentGitHub stars: 40.4k

Agentic Eval skill: what it does and how to install it

Builds iterative evaluation and refinement loops for AI outputs, including self-critique, rubrics, LLM judges and test-driven code fixes.

Summary generated from the skill's documentation.

Install

$ npx skills add github/awesome-copilot --skill agentic-eval

Run it in a terminal. If your agent is already running, start a new session so it picks the skill up.

About this skill

What it does. Provides patterns for generating, evaluating, critiquing and refining outputs in bounded loops. It covers self-reflection, evaluator-optimizer designs, code refinement through tests, rubric scoring and structured JSON evaluation, with convergence checks and iteration logging.

When to use it. For quality-critical code, reports or analysis with clear success criteria, style requirements or compliance standards. It also applies when measuring and improving agent response quality.

History

Repo stars

40.4k+707 since 5 Oct 2026

Stars are counted for the whole repository, which holds 80 skills.

Show as a table
Repo stars by day
DateRepo stars
26 Oct 202640,426
25 Oct 202640,385
24 Oct 202640,365
23 Oct 202640,341
22 Oct 202640,302
21 Oct 202640,260
20 Oct 202640,214
19 Oct 202640,178
18 Oct 202640,143
17 Oct 202640,111
16 Oct 202640,072
15 Oct 202640,038
14 Oct 202639,999
13 Oct 202639,981
12 Oct 202639,948
11 Oct 202639,904
10 Oct 202639,881
9 Oct 202639,845
8 Oct 202639,804
7 Oct 202639,761
6 Oct 202639,731
5 Oct 202639,719

Installs

10.2k+11 since 19 Oct 2026

Installs via skills.sh

Show as a table
Installs by day
DateInstalls
26 Oct 202610,218
25 Oct 202610,215
24 Oct 202610,213
23 Oct 202610,213
22 Oct 202610,210
21 Oct 202610,210
20 Oct 202610,208
19 Oct 202610,207

Similar skills