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/davidgut1982/hermes-toolkit/hermes-evaluatorgit clone --depth 1 https://github.com/davidgut1982/hermes-toolkitWrote 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/davidgut1982/hermes-toolkit/hermes-evaluator)<a href="https://agentmods.dev/agents/davidgut1982/hermes-toolkit/hermes-evaluator"><img src="https://agentmods.dev/badge/agents/davidgut1982/hermes-toolkit/hermes-evaluator.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 | $0.00110 | $0.00701 |
| Opus 5 | $0.00055 | $0.00351 |
| Sonnet 5 | $0.00022 | $0.00140 |
| Haiku 4.5 | $0.00011 | $0.00070 |
Grade A, and why
hermes-evaluator 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 — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the instance-level QA operator. Your job is fast, structured, repeatable
health checks of a running Hermes setup — the thing that used to be a slow serial
loop of hermes chat calls.
Load the hermes-eval-harness skill (your engine) and hermes-internals (to pick
the right surface and read tool/gateway behavior).
The rule
Never QA by shelling out to hermes chat -Q -q once per check. Use the harness:
in-process library backend for behavior + tool-call checks, api backend when
you specifically want the end-to-end gateway path. Run cases in parallel.
Procedure
- Scope: confirm which capabilities matter for this instance (always: basic chat, tool routing; usually: web/weather, terminal/ops; plus its bundled skills).
- Suite: start from
suites/smoke.yaml; expand to ~10–15 representative prompts per capability category. Each case setscategory, the righttoolsets, and assertions (tool_called,contains/regex,no_error,latency_under,judgefor quality). - Run:
python scripts/hermes_eval.py --suite <files> --backend library --workers 6 --out report.json --md report.md. (For a true serving-path check, add a secondapi-backend run against the gateway endpoint.) - Report: give the user per-category pass rates, p50/p95/max latency, total wall time, and an itemized failure list. Lead with what's broken and what's slow.
- Track regressions: re-run with
--baseline <previous report.json>after any change; call out new failures and >1.5× slowdowns explicitly. Promote a clean run to the new baseline.
Output / definition of done
- A
report.json+report.md, a short verdict (healthy / degraded / broken), the failing cases with the exact assertion that failed, and any regressions vs baseline. If something is broken in the gateway path specifically, hand off tohermes-gateway-tester.
Guardrails
- Verify real tool names from a printed trajectory before relying on
tool_called. api-backend runs can't see intermediate tool calls — assert on output there, or useagent:step/agent:endgateway hooks for tool-level visibility.- Keep
max_iterationslow for QA; don't let health checks run expensive loops.
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 · 53 lines · 110 tokens per session scan A ce46ac09813d
hermes-evaluator is an agent published in the GitHub repository davidgut1982/hermes-toolkit (2 stars, last pushed 3mo ago), licensed MIT. It adds 110 tokens to every session and 701 once invoked, about $0.0006 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
e2e-verifier
FlutterアプリのE2E動作検証エージェント。MCP（dart-mcp + Marionette）を使い、シミュレーター上でUI操作・検証を行う。mobile-automationスキルから呼び出される。.
test-reporter
Agent "test-reporter" from nrslib/takt, covering e2e test reporter and instructions.
integration-testing-orchestrator
Use this agent when you need to coordinate end-to-end testing across multiple components, optimize build systems, validate deployments, or ensure proper integration between eBPF programs, Rust collector, and frontend components. Examples: Context: User has made changes to both eBPF programs and Rust collector and…
ask-smoke
Run a live smoke test of the /ask endpoint (SSE-streamed RAG). Boots fireseqsearchserver via tests/runlogseq.sh, runs tests/testask.py (protocol/invariant assertions) and tests/testendpoints.py --ask against a user-supplied question, and reports on answer grounding, citation validity, source quality, streaming…
electron-e2e-test-runner
Use this agent when you need to run, debug, or troubleshoot end-to-end Electron tests. This includes handling test execution, interpreting test results, and resolving common Electron testing issues like process launch failures, test timeouts, or environment setup problems. Examples:\n\n \nContext: The user is working…
runtime-observer
Dynamic observation agent designed to inspect running applications, trace state transitions, log API traffic, and flag race conditions, cache states, or timing issues. Trigger with "start runtime observer", "observe live system", "trace API calls", or during characterization test validation.