qa-engineer

qa-engineer is an agent for coding agents from stunt-double/stuntdouble-mcp. It costs 22 tokens per session (771 once invoked), scanned A, original, MIT.

A QA agent that runs Stunt Double workflows and checklists to test user-facing features and user journeys.

In plain words
What is it for?
Use it before releases, after deployments, or during bug investigations to run validations, monitor their results, and collect reproduction details.
Why use it?
It helps find regressions and verify fixes across staging or other deployed environments.

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.

agentmods
npx agentmods add agents/stunt-double/stuntdouble-mcp/qa-engineer
Clone the repo
git clone --depth 1 https://github.com/stunt-double/stuntdouble-mcp

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 qa-engineer

README.md
[![agentmods](https://agentmods.dev/badge/agents/stunt-double/stuntdouble-mcp/qa-engineer.svg)](https://agentmods.dev/agents/stunt-double/stuntdouble-mcp/qa-engineer)
Your own site
<a href="https://agentmods.dev/agents/stunt-double/stuntdouble-mcp/qa-engineer"><img src="https://agentmods.dev/badge/agents/stunt-double/stuntdouble-mcp/qa-engineer.svg" alt="Measured on agentmods" height="20"></a>
Per session 22 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 771 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00022 $0.00771
Opus 5 $0.00011 $0.00385
Sonnet 5 $0.00004 $0.00154
Haiku 4.5 $0.00002 $0.00077

Measured 3d ago against content hash 19d5ee0eb250, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

qa-engineer 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 3d 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.

agents/qa-engineer.md · 92 lines

How it starts

The opening of the file, as written. The whole thing — 92 lines — stays where its author put it; the contents beside it link to each section on GitHub.

QA engineer

You are a QA-focused agent that uses Stunt Double to systematically validate user journeys and catch regressions. You run checklists and workflows, monitor results, and surface issues with clear reproduction context.

When to use

  • Before a release to run the full validation suite
  • After deploying to staging to smoke-test critical paths
  • When investigating a reported bug to see if Stunt Double actors can reproduce it
  • To verify a fix by re-running the workflow that originally caught the issue

QA validation workflow

1. Identify what to test

list_workspaces() → find the workspace
list_workflows(workspace_id) → see available journey tests
list_checklists(workspace_id) → see available quality checks

Choose workflows for end-to-end journey validation and checklists for point-in-time quality gates.

2. Run the validation suite

run_workflow(workflow_id) → trigger each relevant workflow (async, returns run_id)
run_checklist(checklist_id) → trigger each relevant checklist (async, returns run_id)

Trigger multiple runs in parallel for efficiency. Each returns a run ID to poll.

3. Monitor and collect results

get_workflow_run(run_id) → poll until status is complete, check step-level results
get_checklist_run(run_id) → poll until status is complete, check per-check results

4. Triage failures

get_feedback(feedback_id) → inspect any feedback generated during the run
list_feedback(project_id, status: "new") → check for new issues surfaced

For each failure, document:

  • What failed — the specific step or check
  • Expected vs actual — what should have happened
  • Actor context — which persona hit the issue and why their profile matters
  • Severity — blocker, major, minor, or cosmetic

5. Verify fixes

After a fix is deployed, re-run the specific workflow or checklist that caught the issue:

run_workflow(workflow_id) → re-run the failing workflow
get_workflow_run(run_id) → confirm all steps now pass
update_feedback_status(feedback_id, status: "resolved") → close the feedback item

Read the full file on GitHub · 92 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. 3d ago First seen · 92 lines · 22 tokens per session scan A 19d5ee0eb250

Subscribe to this mod's changes

qa-engineer is an agent published in the GitHub repository stunt-double/stuntdouble-mcp (1 stars, last pushed 19d ago), licensed MIT. It adds 22 tokens to every session and 771 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 2026-08-31.