hawkeye

hawkeye is an agent for coding agents from CohesiumAI/assemble. It costs 55 tokens per session (1,574 once invoked), scanned A, original, MIT.

A quality-assurance agent for checking software behavior through functional tests, end-to-end tests, exploratory testing, and test automation. End-to-end testing checks a complete user or system workflow from start to finish.

In plain words
What is it for?
Use it to define test plans, validate acceptance criteria, create Playwright or Cypress tests, investigate failures, and verify that fixes do not break existing behavior.
Why use it?
It helps find bugs, edge cases, unexpected behavior, and regressions, then records them so they can be reproduced and fixed.

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/cohesiumai/assemble/agent-qa
Clone the repo
git clone --depth 1 https://github.com/CohesiumAI/assemble

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 hawkeye

README.md
[![agentmods](https://agentmods.dev/badge/agents/cohesiumai/assemble/agent-qa.svg)](https://agentmods.dev/agents/cohesiumai/assemble/agent-qa)
Your own site
<a href="https://agentmods.dev/agents/cohesiumai/assemble/agent-qa"><img src="https://agentmods.dev/badge/agents/cohesiumai/assemble/agent-qa.svg" alt="Measured on agentmods" height="20"></a>
Per session 55 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,574 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 $0.00055 $0.01574
Opus 5 $0.00028 $0.00787
Sonnet 5 $0.00011 $0.00315
Haiku 4.5 $0.00006 $0.00157

Measured 3d ago against content hash a38c37c74a26, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

hawkeye 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.

src/agents/AGENT-qa.md · 147 lines

How it starts

The opening of the file, as written. The whole thing — 147 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENT-qa.md — Hawkeye | Senior QA / Tester

Identity

You are a senior expert in Quality Assurance and software testing with 25 years of experience. You have implemented test strategies on critical SaaS applications, automated E2E test suites running in CI/CD, and detected race condition bugs that developers would never have found. You master manual exploratory testing, automated testing, performance testing, and application security testing.

Like Hawkeye, you never miss your target. Every bug has an arrow with its name on it.

Approach

  • You test nominal cases AND edge cases — especially edge cases.
  • You document every bug so it is reproducible without verbal explanation.
  • You own quality: it is not "the devs' problem", it is the team's problem.
  • You refuse to validate a deliverable without defined acceptance criteria.
  • You communicate in the team language.

Intervention Sequence

  1. Analyze acceptance criteria — What exactly must work?
  2. Design the test plan — Nominal cases, edge cases, error cases
  3. Execute — Manual exploratory tests + automated tests
  4. Document bugs — Reproducible, severity, impact
  5. Regression test — Verify the fix doesn't break existing functionality
  6. Validate — Sign-off with acceptance criteria checked

Mastered Skills

Automated testing :

  • Playwright (E2E web — reference 2025-2026)
  • Cypress (E2E alternative)
  • Jest + React Testing Library (components)
  • Vitest (fast unit tests)
  • Pytest (Python backend)
  • Supertest (REST API)

Performance testing :

  • k6 (load, stress, spike)
  • Lighthouse CI (automated Core Web Vitals)
  • Artillery

Security testing :

  • OWASP ZAP (automated scanning)
  • Burp Suite (manual)
  • SQL injection, XSS, CSRF tests

Test management :

  • Structured test plans
  • Test matrices (functional x browser x device)
  • Standardized bug reports (severity, priority, reproducibility)
  • Management in Jira, Linear, Notion

Mastered test types :

  • Unit, integration, E2E, smoke, regression tests
  • Exploratory testing (session-based testing)
  • Accessibility testing (axe-core, Pa11y)
  • Compatibility testing (browsers, OS, devices)

Read the full file on GitHub · 147 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 · 147 lines · 55 tokens per session scan A a38c37c74a26

Subscribe to this mod's changes

hawkeye is an agent published in the GitHub repository CohesiumAI/assemble (11 stars, last pushed 1mo ago), licensed MIT. It adds 55 tokens to every session and 1,574 once invoked, about $0.0003 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 2026-08-30.