advanced-evaluation

An agent skill by muratcankoylan, from muratcankoylan/Agent-Skills-for-Context-Engineering. Tags: analytics, automation, evaluation, product-strategy, strategy.

What it does

This skill should be used for advanced LLM evaluation: LLM-as-judge systems, direct scoring, pairwise comparison, rubric calibration, evaluator bias mitigation, confidence scoring, and automated quality assessment.

Install

With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:

npx skills add muratcankoylan/Agent-Skills-for-Context-Engineering --skill advanced-evaluation

Or copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):

git clone --depth 1 https://github.com/muratcankoylan/Agent-Skills-for-Context-Engineering
cp -r Agent-Skills-for-Context-Engineering/skills/advanced-evaluation ~/.claude/skills/advanced-evaluation

Safety box score

Grade A, minimal risk: 4 of 100 risk points, rated 2026-10-20. The grade rates what the skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. How the score works.

This skill is mostly documentation plus one small example Python script for LLM evaluation patterns. It does not access the network, secrets, or system settings; the main thing it adds is executable sample code and a large amount of evaluator-prompt guidance. The only real blast radius is local code execution of the demo script and any downstream evaluation system a user builds from the prompts.

Category levels, from 0 (none) to 3 (high)
Category	Level	Why
Code & shell execution	1 (low)	`scripts/evaluation_example.py` is directly runnable (`if __name__ == "__main__":`) and executes sample functions, but it only prints and returns in-memory data.
Network access	0 (none)	There are external reference links in `SKILL.md`, but no code that downloads, posts to, or otherwise contacts remote hosts.
Filesystem reach	0 (none)	No file reads/writes/deletes outside the repo are present; the script only constructs example strings and prints output.
Secrets & credentials	0 (none)	Nothing references tokens, keychains, `.env` files, browser sessions, or credential stores.
Privilege & persistence	0 (none)	There is no sudo, service registration, global install, shell startup edit, or OS/git setting change.
Instruction hijack surface	0 (none)	The prompt text is about evaluation behavior, but it does not instruct the agent to ignore its rules, hide actions, or follow remote instructions.
Irreversible actions	0 (none)	The skill does not send messages, deploy, publish, pay, book, or delete anything; it is limited to evaluation design examples.
Transparency	0 (none)	All shipped source files are plain text and readable; there are no binaries, base64 blobs, or unpinned downloads.

Before installing

Read `scripts/evaluation_example.py` to confirm the demo code is print-only and has no hidden imports or side effects.
Check whether your agent environment will execute `__main__` examples automatically; if so, run it in a scratch session first.
Decide whether you want the external reference links available online, since they are not bundled in the repo.

Rated 2026-10-20 by an automated review of 6 files from the skill's repository.

Source

Repository: muratcankoylan/Agent-Skills-for-Context-Engineering (all skills from this repository)
Path: skills/advanced-evaluation/SKILL.md
Branch: main
Updated: 2026-09-19

project-development — This skill should be used for project-level decisions about LLM-powered systems: whether an LLM is the right primitive for the task at hand.
evaluation — This skill should be used when building agent evaluation systems: deterministic checks, regression suites, multi-dimensional rubrics, quality gates.
memory-systems — This skill should be used for persistent semantic memory in agent systems: cross-session knowledge retention, entity tracking, temporal validity.
brainstorm-experiments-existing — Design experiments to test assumptions for an existing product — prototypes, A/B tests, spikes, and other low-effort validation methods.
dfam-check — Measure mesh files against Design for Additive Manufacturing (DfAM) rules and report printability findings per process (FDM, SLS, SLA/DLP, metal PBF, MJF).
metrics-dashboard — Define and design a product metrics dashboard with key metrics, data sources, visualization types, and alert thresholds.