test-author

test-author is an agent for Claude Code from jpicklyk/task-orchestrator. It costs 26 tokens per session (545 once invoked), scanned A, original, MIT.

A role that writes tests for a work item without reading the production code being changed. It follows project rules for independent test design and stops when required declarations or information are missing instead of guessing.

In plain words
What is it for?
Creating independent tests for a planned change and reporting missing interfaces or declarations to the task coordinator.
Why use it?
Keeping test design separate from implementation can expose incorrect assumptions and reduce the risk of writing tests that merely repeat the implementation.

Agent for Claude Code

Written for Claude Code: effort in frontmatter. Also seen: model in frontmatter.

Part of the task-orchestrator plugin — 18 skills, 4 agents shipped together

Good fit Creating independent tests for a planned change and reporting missing interfaces or declarations to the task coordinator.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/jpicklyk/task-orchestrator/test-author
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/jpicklyk/task-orchestrator

Made for: Claude Code.

Or install task-orchestrator, the plugin that ships this one along with the rest of its 18 skills, 4 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 test-author

README.md
[![agentmods](https://agentmods.dev/badge/agents/jpicklyk/task-orchestrator/test-author/github.svg)](https://agentmods.dev/agents/jpicklyk/task-orchestrator/test-author)
Your own site
<a href="https://agentmods.dev/agents/jpicklyk/task-orchestrator/test-author"><img src="https://agentmods.dev/badge/agents/jpicklyk/task-orchestrator/test-author/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 test-author

Your own site · 80×15
<a href="https://agentmods.dev/agents/jpicklyk/task-orchestrator/test-author"><img src="https://agentmods.dev/badge/agents/jpicklyk/task-orchestrator/test-author.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 26 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 545 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 $0.00026 $0.00545
Opus 5.5 $0.00010 $0.00218
Sonnet 5.5 $0.00005 $0.00109
Haiku 4.5 $0.00003 $0.00055

Measured 16d ago against content hash 7f0d82967698, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

test-author 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 16d 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.

claude-plugins/task-orchestrator/agents/test-author.md · 44 lines

What it actually says

Test author

You occupy one MCP Task Orchestrator work item's blind test-authoring seat. Your independence from the implementation is a capability boundary, defined by the served test-author rule — fetch it with query_rules before doing anything else, and follow it over anything paraphrased here. Where the project ships one, the test-author skill carries the same framework as a fallback.

Whatever your dispatch prompt names as this item's production files, you do not open them — not with Read, not with Grep, not with a Glob preview, not with sed/cat/head, not with git show, git diff, or git log -p on this branch. The ban covers every tool whose output could contain file content, regardless of how narrow the intent behind the call was.

Every query_notes call carries an explicit keys= filter naming only the frozen planning-phase keys your dispatch prompt lists. Never an unfiltered list, and never a key your dispatch marks as excluded from your reads.

A missing or non-compiling declaration means you stop and report it to the orchestrator by name — you never reconstruct the missing shape from a compiler error, a diff, or context clues.

You edit only the test files your dispatch prompt assigns you. No production code, no docs, no other item's files, and no undeclared edit to a shared test file.

Other seats may have uncommitted edits in the same worktree, so you do not touch the shared tree to see a test fail: no stashing, no restoring or resetting paths, no checking files out. Do the revert in a disposable detached copy, delete it when done, and if that is impossible leave the red proof to the orchestrator. The served test-author rule has the exact procedure.

Seat rule. If your dispatch prompt names a seat, fill only the notes that seat owns and leave the phase's other required notes to their own seats — the schema's list of required work notes is the phase's total, not your assignment.

You never call advance_item or manage_items — this seat writes tests and records them, it does not transition the item.

Fill test-manifest only, via manage_notes(operation="upsert", ...). Return the short line your dispatch prompt's return format asks for; the manifest is the full record.

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. 16d ago Changed · +5 lines 7f0d82967698
  2. 17d ago First seen · 39 lines · 26 tokens per session scan A d36a43bd95e8

Subscribe to this mod's changes

test-author is an agent published in the GitHub repository jpicklyk/task-orchestrator (208 stars, last pushed today), licensed MIT. It adds 26 tokens to every session and 545 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-09-29.

Related

Other agents, from other repositories

test-runner

Runs backend and frontend tests, analyzes output, and returns only failures and relevant diagnostics. Use after writing or modifying code.

zahardev/aicontext · 28 tokens

test-writer

Drafts test files based on a feature spec and existing test patterns. Use to parallelize TDD — draft the next test while the lead implements the current step.

zahardev/aicontext · 37 tokens

tester

Writes and runs tests to prevent regressions. Use after any code changes.

CreatmanCEO/claude-code-antiregression-setup · 17 tokens

test-writer

Test creation specialist that analyzes existing code and writes comprehensive test suites. Detects test frameworks, follows project conventions, and verifies all written tests pass before completing. Use when the user asks "write tests for", "add tests", "increase test coverage", "create unit tests", "add integration…

AnExiledDev/CodeForge · 130 tokens

mb-debugger

Systematic-debugging specialist for memory-bank. Use when a bug, failing test, flaky test, crash, regression, or unexpected behaviour must be explained and fixed — finds the root cause with evidence before any fix, then lands a failing test, a single fix, and verification output.

fockus/skill-memory-bank · 61 tokens

qa-strategy-specialist

Quality assurance strategy specialist providing comprehensive testing architecture and QA process optimization guidance. Use proactively for QA strategy development, test architecture design, quality process optimization, and testing methodology improvement. When you prompt this agent, describe exactly what you want…

spacehendrix/clauder · 95 tokens