red-test-evaluator

red-test-evaluator is an agent for Claude Code from navapbc/digital-service-orchestra. It costs 32 tokens per session (4,647 once invoked), scanned A, original, Apache-2.0.

An evaluation agent for failed RED test-writing tasks. RED tests are tests that initially fail because they describe behavior the code does not yet provide; the agent returns REVISE, REJECT, or CONFIRM.

In plain words
What is it for?
Checking test-writer failures against a shared behavioral testing standard and triaging the result with one of three verdicts.
Why use it?
It helps determine whether a failed test-writing attempt needs correction, should be discarded, or correctly represents the intended behavior.

Agent for Claude Code

Written for Claude Code: ${CLAUDE_PLUGIN_ROOT} variable. Also seen: model in frontmatter; mentions subagents; mentions Claude Code.

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the dso plugin — 37 skills, 4 commands, 53 agents shipped together

Good fit Checking test-writer failures against a shared behavioral testing standard and triaging the result with one of three verdicts.

Compare 6 agents from other repositories ↓
Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add navapbc/digital-service-orchestra
Claude Code
/plugin install dso

Made for: Claude Code.

Or install dso, the plugin that ships this one along with the rest of its 37 skills, 4 commands, 53 agents.

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 red-test-evaluator

README.md
[![agentmods](https://agentmods.dev/badge/agents/navapbc/digital-service-orchestra/red-test-evaluator/github.svg)](https://agentmods.dev/agents/navapbc/digital-service-orchestra/red-test-evaluator)
Your own site
<a href="https://agentmods.dev/agents/navapbc/digital-service-orchestra/red-test-evaluator"><img src="https://agentmods.dev/badge/agents/navapbc/digital-service-orchestra/red-test-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 red-test-evaluator

Your own site · 80×15
<a href="https://agentmods.dev/agents/navapbc/digital-service-orchestra/red-test-evaluator"><img src="https://agentmods.dev/badge/agents/navapbc/digital-service-orchestra/red-test-evaluator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 32 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 4,647 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 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 $0.00032 $0.04647
Opus 5 $0.00016 $0.02323
Sonnet 5 $0.00006 $0.00929
Haiku 4.5 $0.00003 $0.00465

Measured 12d ago against content hash 5cc8b867f4b3, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

red-test-evaluator scanned grade A 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 12d 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.

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

1. **Check `SUGGESTED_ALTERNATIVE`**: If the writer's rejection payload includes a `SUGGESTED_ALTERNATIVE` that points to a valid restructuring path (e.g., mock `subprocess.run`, patch `time.sleep`, use small fixture dat
plugins/dso/agents/red-test-evaluator.md · 411 lines

How it starts

The opening of the file, as written. The whole thing — 411 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Red Test Evaluator

Startup: Session HEAD Sync (worktree isolation fix)

When dispatched with isolation: "worktree", the Agent runtime creates your worktree branched from origin/main — NOT from the orchestrator's session HEAD. If the orchestrator injected SESSION_BRANCH and SESSION_HEAD into your prompt, sync to the session HEAD as your FIRST action before reading any source files. Bug a951-d6f2-0c21-443f tracks this.

if [[ -n "${SESSION_BRANCH:-}" && -n "${SESSION_HEAD:-}" ]]; then
    bash "${CLAUDE_PLUGIN_ROOT}/scripts/worktree-session-head-sync.sh"  # shim-exempt: internal orchestration script
    if [[ $? -ne 0 ]]; then
        echo "ERROR: worktree-session-head-sync.sh failed — aborting" >&2
        exit 1
    fi
elif [[ -n "${SESSION_BRANCH:-}" || -n "${SESSION_HEAD:-}" ]]; then
    echo "WARNING: SESSION_BRANCH/SESSION_HEAD partially set — skipping worktree sync" >&2
fi

When both are unset (orchestrator on main, no session in flight), do nothing — your default origin/main worktree is correct.

Section 1: Role and Identity

You are an opus-tier triage agent for RED test writer failures. When the dso:red-test-writer cannot produce a behavioral test and emits TEST_RESULT:rejected, you receive that rejection payload along with an orchestrator context envelope and determine the correct verdict.

Your job is to answer: Is this rejection legitimate, fixable, or out of scope for automated TDD?

You return exactly one verdict per invocation:

  • REVISE — the task list or description needs adjustment; a behavioral test is achievable after revision
  • REJECT — the task is outside automated TDD scope; no revision would yield a testable assertion
  • CONFIRM — the rejection is legitimate; TDD is genuinely infeasible for this task type

Read the full file on GitHub · 411 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. 12d ago First seen · 411 lines · 32 tokens per session scan A 5cc8b867f4b3

Subscribe to this mod's changes

red-test-evaluator is an agent published in the GitHub repository navapbc/digital-service-orchestra (6 stars, last pushed today), licensed Apache-2.0. It adds 32 tokens to every session and 4,647 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.