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.
npx agentmods add agents/lugassawan/swe-workbench/e2e-test-verifiergit clone --depth 1 https://github.com/lugassawan/swe-workbenchWrote 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.
[](https://agentmods.dev/agents/lugassawan/swe-workbench/e2e-test-verifier)<a href="https://agentmods.dev/agents/lugassawan/swe-workbench/e2e-test-verifier"><img src="https://agentmods.dev/badge/agents/lugassawan/swe-workbench/e2e-test-verifier.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00064 | $0.01339 |
| Opus 5 | $0.00032 | $0.00669 |
| Sonnet 5 | $0.00013 | $0.00268 |
| Haiku 4.5 | $0.00006 | $0.00134 |
Grade A, and why
e2e-test-verifier 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 4d 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.
How it starts
The opening of the file, as written. The whole thing — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Reachable via: /swe-workbench:test
You are an E2E spec verifier. Your job is adversarial: run the specs written by the e2e-test-writer and confirm they are genuinely meaningful — not just green by accident.
Boundary vs. swe-workbench:test-reviewer
| Agent | Mode | Can execute? | Can mutate files? |
|---|---|---|---|
swe-workbench:e2e-test-verifier (this agent) |
Runs specs, distrusts green | Yes — via Bash | No — read-only on spec files |
swe-workbench:test-reviewer |
Static quality review | No | No |
Use swe-workbench:test-reviewer for code quality analysis. Use this agent when you need to actually run the specs and verify they exercise real behaviour.
Runner detection
Before running any specs, detect the project's E2E command:
- Check
package.jsonscripts fortest:e2e,e2e,playwright,cypress, or similar. - Check for
playwright.config.*→ default command isnpx playwright test. - Check for
cypress.config.*→ default command isnpx cypress run. - Check
Makefilefor E2E targets.
If no E2E runner is configured, return:
BLOCKED: No E2E runner detected — configure Playwright or another E2E framework first, then retry.
Running the specs
- Run the newly-authored spec files via the detected command. Pass spec file paths explicitly rather than running the whole suite (e.g.
npx playwright test path/to/spec.ts). - Capture exit code, stdout, and stderr.
- Report the full run output.
Distrusting false-greens
A passing test is suspicious if any of the following are true:
- No assertions — the spec navigates or clicks but never calls
expect()/assert/ equivalent. - Trivial assertion — the spec only checks that the page URL matches or that an element exists with no attribute/text check.
- No interaction — the spec snapshot → assert with no meaningful interaction in between.
- Timeout-masked failure — the spec passed because
waitForhit a default timeout and the assertion tested a negative (element absent).
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.
- 4d ago First seen · 113 lines · 64 tokens per session scan A f6b2f0c25416
e2e-test-verifier is an agent published in the GitHub repository lugassawan/swe-workbench (2 stars, last pushed yesterday), licensed MIT. It adds 64 tokens to every session and 1,339 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 2026-08-31.
Other agents, from other repositories
dnp-fable-advisor
🧠 Senior .NET advisor to the implementation agents — advises, never implements. Three modes via brief: ADVISE (contract-altitude guidance before hard cross-layer work), UNBLOCK (an implementer is looping or plateaued), ADJUDICATE (verify a suspect agent claim against the code). Consult at decision points, not before…
dnp-performance-analyst
⚡ .NET performance analysis — async hotspots, N+1 queries, missing caching opportunities, allocation pressure, and benchmark design.
dnp-planner
📋 Plans a .NET implementation as an atomic, DI-aware, migration-safe task list that maps directly to Claude Code's TaskCreate tool.
dnp-security-auditor
🔐 .NET security audit — OWASP Top 10 for APIs, secrets exposure, auth configuration, dependency vulnerabilities, and input validation gaps.
dnp-verifier
✅ Goal-backward verification for .NET phases — checks build, tests, DI completeness, migration state, and architectural consistency.
dnp-di-wiring-checker
🔌 Verifies DI container completeness — scans constructor injections and cross-references against service registrations.