An evaluation workflow for defining checks before coding, running them, and tracking whether a feature works. It records both new capability checks and checks that protect existing behavior.

Use it to create evaluation files, verify feature and regression criteria, and view current pass/fail status.

It makes success criteria explicit before implementation and shows which requirements or regressions still fail.

ccc-eval — Command for Claude Code by KevinZai/commander

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/KevinZai/commander

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.

README.md

[![agentmods](https://agentmods.dev/badge/commands/kevinzai/commander/ccc-eval/github.svg)](https://agentmods.dev/commands/kevinzai/commander/ccc-eval)

Your own site

<a href="https://agentmods.dev/commands/kevinzai/commander/ccc-eval"><img src="https://agentmods.dev/badge/commands/kevinzai/commander/ccc-eval/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.

Your own site · 80×15

<a href="https://agentmods.dev/commands/kevinzai/commander/ccc-eval"><img src="https://agentmods.dev/badge/commands/kevinzai/commander/ccc-eval.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 45 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,299 The whole file, excluding the scripts and references it only reads on demand.

Security scan B 1 finding. A grade says what 26 rules found in the file — not that it is safe.

Origin 89% copy Near-identical to another mod 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.

Model	Per session	Once invoked
Fable 5.1	$0.00045	$0.01299
Opus 5.5	$0.00018	$0.00520
Sonnet 5.5	$0.00009	$0.00260
Haiku 4.5	$0.00005	$0.00130

Measured 5d ago against content hash 1ffb47fc1f77, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade B, and why

ccc-eval scanned grade B with 1 finding 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 5d 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.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

4. Verify: `test -f .claude/evals/<name>.md && grep -c '^- \[ \]' .claude/evals/<name>.md` prints the number of criteria.

Origin

This is a copy

89% identical to eval — 188 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

commands/ccc-eval.md · 135 lines

How it starts

The opening of the file, as written. The whole thing — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/ccc-eval — eval-driven feature tracking

CC Commander · /ccc-eval · eval definitions, run logs, and a ship/no-ship report per feature.

Arguments

$ARGUMENTS:

define <name> — create a new eval definition
check <name> — run the evals and record results
report <name> — produce the full report
list — show every eval definition and its status
clean [<name>] — trim old run logs, keeping the last 10 runs

No argument behaves like list. A missing <name> for define, check or report: ask for it in plain text.

What happens

define

If .claude/evals/<name>.md already exists, show it and stop; never overwrite a definition.

Run mkdir -p .claude/evals and write .claude/evals/<name>.md with today's date in place of <date>:

## EVAL: <name>
Created: <date>

### Capability Evals
- [ ] [Description of capability 1]
- [ ] [Description of capability 2]

### Regression Evals
- [ ] [Existing behavior 1 still works]
- [ ] [Existing behavior 2 still works]

### Success Criteria
- pass@3 > 90% for capability evals
- pass^3 = 100% for regression evals

Ask the user to replace each bracketed placeholder with a concrete, checkable criterion (an observable output, a command result, or a test name). Write their answers into the file.
Verify: test -f .claude/evals/<name>.md && grep -c '^- \[ \]' .claude/evals/<name>.md prints the number of criteria.

check

Read .claude/evals/<name>.md. If it is missing, say so and suggest /ccc-eval define <name>.
Find the project's test command: the test script in package.json (npm test), pytest for pyproject.toml/pytest.ini, go test ./... for go.mod, cargo test for Cargo.toml. If none is found, ask the user for the command.
For each capability eval, verify the criterion directly (run the named command, call the endpoint, read the output). Record PASS or FAIL and the attempt number. A criterion that cannot be checked is FAIL with the reason, never an assumed PASS.
For each regression eval, run the relevant tests with the command from step 2 and compare against the previous run in the log. Record PASS or FAIL.
Append one run entry to .claude/evals/<name>.log: timestamp, each eval id with PASS/FAIL and attempt, and the exact commands run.
Report:

Read the full file on GitHub · 135 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.

5d ago Changed · +10 lines · +26 tokens per session scan A → B 1ffb47fc1f77
1mo ago First seen · 125 lines · 19 tokens per session scan A 438e511fa60e

Subscribe to this mod's changes

ccc-eval is a command published in the GitHub repository KevinZai/commander (6 stars, last pushed today), licensed MIT. It adds 45 tokens to every session and 1,299 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it B with 1 finding (reads agent configuration directories). It is 89% identical to eval, differing in 188 lines, and is treated as a copy.

Claude Code TDD Testing ai-coding anthropic anthropic-claude claude claude-code-plugin

ccc-eval

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade B, and why

Reads agent configuration directoriesmediumAgent snooping

This is a copy

How it starts

/ccc-eval — eval-driven feature tracking

Arguments

What happens

define

check

What this file has done since we first saw it

Other commands, from other repositories

loop

execute

symfony-tdd-pest

symfony-tdd-phpunit

dev-tdd

work-flow-feature