calibration-evaluator

calibration-evaluator is an agent for Claude Code from odere-pro/claude-calibration. It costs 174 tokens per session (2,159 once invoked), scanned B, original, MIT.

An agent that audits Claude Code settings against a calibration checklist and writes reports without editing the settings.

In plain words
What is it for?
It helps evaluate user, project, and plugin settings, compare baseline and follow-up results, and produce per-feature and interaction reports.
Why use it?
It shows configuration problems and changes between an initial review and a later review while keeping the setup untouched.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter; mentions subagents; names the TodoWrite tool.

Part of the claude-calibration plugin — 20 skills, 6 agents, 2 hooks shipped together

Good fit It helps evaluate user, project, and plugin settings, compare baseline and follow-up results, and produce per-feature and interaction reports.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/odere-pro/claude-calibration/calibration-evaluator
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/odere-pro/claude-calibration

Made for: Claude Code.

Or install claude-calibration, the plugin that ships this one along with the rest of its 20 skills, 6 agents, 2 hooks.

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 calibration-evaluator

README.md
[![agentmods](https://agentmods.dev/badge/agents/odere-pro/claude-calibration/calibration-evaluator/github.svg)](https://agentmods.dev/agents/odere-pro/claude-calibration/calibration-evaluator)
Your own site
<a href="https://agentmods.dev/agents/odere-pro/claude-calibration/calibration-evaluator"><img src="https://agentmods.dev/badge/agents/odere-pro/claude-calibration/calibration-evaluator/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 calibration-evaluator

Your own site · 80×15
<a href="https://agentmods.dev/agents/odere-pro/claude-calibration/calibration-evaluator"><img src="https://agentmods.dev/badge/agents/odere-pro/claude-calibration/calibration-evaluator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 174 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,159 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 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 $0.00174 $0.02159
Opus 5 $0.00087 $0.01079
Sonnet 5 $0.00035 $0.00432
Haiku 4.5 $0.00017 $0.00216

Measured 8d ago against content hash 9438a695f0be, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade B, and why

calibration-evaluator 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 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.

Recursive force deletemediumDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

verbatim — they're already slim. Then `rm -rf <Run folder>/.drafts`.

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

agents/calibration-evaluator.md · 143 lines

How it starts

The opening of the file, as written. The whole thing — 143 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the calibration evaluator. You audit the setup and write reports; you never edit Claude Code config. Your only writes are to <Run folder>/eval-*.md and the plan.md frontmatter fields you own (last_phase_completed, baseline_severity, baseline_reports, last_evaluation).

Inputs (in the spawn prompt)

Pass: 1 (baseline) or 2 (delta) · Run folder: absolute path · Plan: <run>/plan.md · Rubric dir: absolute path to docs/ (fallback) · Bundles dir: absolute path to <plugin>/skills/ (primary — for each feature read <Bundles dir>/calibrate-<feature>/reference.md for the rubric and run that bundle's scripts/enumerate.sh + scripts/lint.sh for actual numbers) · Project dir: absolute path · Audit scope: user + project + plugins · Plugin filter: the canonical include:…|exclude:…|scope:… spec (or empty = all plugins) — pass it verbatim to every calibration-feature-evaluator you spawn (both passes use the same value). For Pass 2 only: Baseline reports: comma-separated filenames of the Pass-1 reports under <Run folder>.

If Bundles dir is UNKNOWN or empty, fall back to <Rubric dir>/features/<feature>.md and use signature names from <Bundles dir>/../rules/signatures.md. If both unreachable, derive sensible signatures from the rubric prose but flag the report so the recurrence detector knows it may underperform.

Pass 1 — baseline

Parallel fan-out to 9 calibration-feature-evaluator subagents (features: claude-md, rules, settings, skills, subagents, hooks, mcp, plugins, general), then sequential cross-feature work.

  1. mkdir -p <Run folder>/.drafts.

  2. Fan out 9 feature evaluators in one tool-use block. Spawn prompt per feature:

    Agent(calibration-feature-evaluator)
    Pass: 1 (baseline).
    Feature: <feature>.
    Run folder: <Run folder>.
    Bundles dir: <Bundles dir>.
    Rubric dir: <Rubric dir>.
    Project dir: <Project dir>.
    Plugin filter: <Plugin filter>.
    Draft path: <Run folder>/.drafts/feat-<feature>.md.
    

Read the full file on GitHub · 143 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 · 143 lines · 174 tokens per session scan B 9438a695f0be

Subscribe to this mod's changes

calibration-evaluator is an agent published in the GitHub repository odere-pro/claude-calibration (1 stars, last pushed 1mo ago), licensed MIT. It adds 174 tokens to every session and 2,159 once invoked, about $0.0009 per session on Opus 5. A static security scan graded it B with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other agents, from other repositories

qa-engineer

Writes cross-slice E2E tests covering flows that span multiple slices. Runs once after all feature slices complete in the Build phase. Reads implementation-log.md to understand the full shape of what was built. Flags which slices caused any E2E failures for targeted retry.

saitarrun/Devforge-ai · 58 tokens

qa-engineer

Test author / runner persona. Authors per-acceptance-criterion test plans and the tests that realize them, then runs the verify/test recipes and reports pass/fail evidence against the contract. Dispatch it to build and execute the test coverage for an authorized atom. It writes and runs tests in a confined tree; it…

lukasrepublic/agentic-foundry · 85 tokens

g-forge-dev

Use proactively to run this repo's test suites and g-dev/ gate fixtures and return attested runner output. The execution counterpart to test-writer, which has no execution tool and cannot run anything — g-forge-dev closes that gap for THIS repo's own dogfooded plugin code only.

onlygian/G-Forge · 64 tokens

test-writer

Use proactively after implementing code that needs coverage or when spec-writer produces a spec. Writes unit, integration, or e2e tests. Fixed data only, no Date.now() or random values.

onlygian/G-Forge · 44 tokens

tester

Writes component, integration, and E2E tests. All non-unit tests run via Testkube in Kubernetes.

LiorCohen/sdd · 25 tokens

edge-case-explorer

Systematically discovers and catalogs edge cases that should be covered by tests for a given piece of code. Traces input sources, call chains, and integration boundaries to find boundary values, type coercion traps, external input messiness, state-dependent failures, and error propagation gaps. Use when exploring how…

testdouble/han · 135 tokens