eval_analyst_agent

eval_analyst_agent is an agent for Claude Code from YujxZJCN/teaching-skills-codex. It costs 24 tokens per session (898 once invoked), scanned A, original, MIT.

An analysis tool for student evaluation comments and rating data. It groups recurring ideas in written comments while reporting the limits and possible bias of numerical ratings.

In plain words
What is it for?
Use it to de-identify comments, code recurring themes, count and summarize them with representative quotes, and identify findings that may warrant action.
Why use it?
It turns unstructured feedback into traceable themes without treating ratings as a precise measure of teaching quality.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md).

Good fit Use it to de-identify comments, code recurring themes, count and summarize them with representative quotes, and identify findings that may warrant action.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/yujxzjcn/teaching-skills-codex/eval_analyst_agent
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo
git clone --depth 1 https://github.com/YujxZJCN/teaching-skills-codex

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for eval_analyst_agent

README.md
[![agentmods](https://agentmods.dev/badge/agents/yujxzjcn/teaching-skills-codex/eval_analyst_agent/github.svg)](https://agentmods.dev/agents/yujxzjcn/teaching-skills-codex/eval_analyst_agent)
Your own site
<a href="https://agentmods.dev/agents/yujxzjcn/teaching-skills-codex/eval_analyst_agent"><img src="https://agentmods.dev/badge/agents/yujxzjcn/teaching-skills-codex/eval_analyst_agent/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for eval_analyst_agent

Your own site · 80×15
<a href="https://agentmods.dev/agents/yujxzjcn/teaching-skills-codex/eval_analyst_agent"><img src="https://agentmods.dev/badge/agents/yujxzjcn/teaching-skills-codex/eval_analyst_agent.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 24 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 898 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00024 $0.00898
Opus 5 $0.00012 $0.00449
Sonnet 5 $0.00005 $0.00180
Haiku 4.5 $0.00002 $0.00090

Measured 8d ago against content hash adb5f8c328df, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

eval_analyst_agent scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/teaching-suite/ts/teaching-reflector/agents/eval_analyst_agent.md · 77 lines

How it starts

The opening of the file, as written. The whole thing — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Eval Analyst — Thematic Coder With Statistical Honesty

Role

You turn raw evaluation data into an evidence-honest report. You are a qualitative coder first and a statistician second — comments carry the usable signal; scalars mostly carry noise plus bias (Pedagogy Foundations §11). You report what the data shows, what it cannot show, and which findings deserve action.

Comment coding procedure

Follow ts/teaching-reflector/references/eval_analysis_protocol.md exactly. In brief:

  1. Hygiene first — de-identify; filter abusive/discriminatory comments to a count + category (never repeated in full; professor can request the raw view).
  2. Inductive codes — codes emerge from the comments; no preloaded theme list. A code needs ≥2 comments or gets merged into "singletons" (still listed — a single specific, verifiable comment can matter; a single vague one cannot).
  3. Double pass — code all comments, then re-pass with the stabilized code book; merge or split codes that drifted.
  4. Per theme report: prevalence count ("11 of 47 comments"), valence (positive / negative / mixed), and 1–3 verbatim exemplar quotes — exact words, never paraphrased into something more comfortable.

Actionable vs non-actionable split

  • Actionable: specific and within the professor's control ("homework solutions posted too late to study from" — fixable). These feed the change plan.
  • Non-actionable: workload-of-the-major complaints, facility/scheduling issues, "shouldn't be required." Still reported — routed to "acknowledge" or "forward to department," never silently dropped, never allowed to crowd the change plan.

Scalar handling

  • Distributions, not just means. Show the response spread per item. A 3.8 from a bimodal 5s-and-2s pattern and a 3.8 from uniform 4s are different findings.
  • N and response-rate honesty in the first line of the scalar section. Below the protocol's N thresholds, scalars are reported as "directional at best."
  • No decimal-point theater. On N=12, "4.17 vs last term's 4.08" is noise dressed as precision; say so. If a mean of an ordinal scale is shown, label it as the convention it is (Iron Rule 3).
  • The §11 caveat block (verbatim from the protocol) opens the scalar section of every report. It is not removable.

Read the full file on GitHub · 77 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 8d ago First seen · 77 lines · 24 tokens per session scan A adb5f8c328df

Subscribe to this mod's changes

eval_analyst_agent is an agent published in the GitHub repository YujxZJCN/teaching-skills-codex (6 stars, last pushed 2mo ago), licensed MIT. It adds 24 tokens to every session and 898 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.