A set of instructions for creating a test harness for SysMoBench, a system used to evaluate software behaviour against specifications. The harness copies a system, instruments it to emit newline-delimited event traces, and provides a script to run it.

Use it when preparing a new evaluation task: clone the source, add trace instrumentation, configure the task paths, write the run script, and document the setup.

It creates the repeatable inputs and recorded behaviour needed to compare an implementation with the required specification.

harness-gen — Skill for Claude Code and Codex by specula-org

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.

Any agent

npx skills add specula-org/SysMoBench --skill harness-gen

Clone the repo

git clone --depth 1 https://github.com/specula-org/SysMoBench

Made for: Claude Code, Codex.

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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/specula-org/sysmobench/harness-gen/github.svg)](https://agentmods.dev/skills/specula-org/sysmobench/harness-gen)

Your own site

<a href="https://agentmods.dev/skills/specula-org/sysmobench/harness-gen"><img src="https://agentmods.dev/badge/skills/specula-org/sysmobench/harness-gen/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/specula-org/sysmobench/harness-gen"><img src="https://agentmods.dev/badge/skills/specula-org/sysmobench/harness-gen.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 79 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 353 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.

Model	Per session	Once invoked
Fable 5.1	$0.00079	$0.00353
Opus 5.5	$0.00032	$0.00141
Sonnet 5.5	$0.00016	$0.00071
Haiku 4.5	$0.00008	$0.00035

Measured 2mo ago against content hash b35601b2f2ac, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

harness-gen 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 2mo ago.

The scan reads SKILL.md. This mod also ships 3 executable files (examples/cometbft/preprocess_trace.py, examples/cometbft/run.sh, examples/dpdk-ring/run.sh), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

tla_eval/skills/harness-gen/SKILL.md · 21 lines

What it actually says

Read guide.md for the full workflow methodology.

SysMoBench conventions (overrides Specula paths in guide.md)

Cloned source lives at artifacts/<task>/ (see artifacts/README.md).
The task's prompt at tla_eval/tasks/<task>/prompts/ defines what must be modeled — the harness must emit events for each required behavior at the right granularity.
After instrumenting, update tla_eval/tasks/<task>/task.yaml's tv.repo_path to point at the cloned tree, and fill tv.harness (instrumentation_file, test_file, run_command).
Trace output goes to artifacts/<task>/traces/ by default (or a path defined by tv.harness.traces_output_env).
Downstream consumer: the tv-eval skill will re-run this harness when spec evaluation needs fresh/compliant traces.

Orchestration: bootstrap the 9 missing tasks

If you were handed this skill as part of the SysMoBench harness-bootstrap project, read BOOTSTRAP_9_SYSTEMS.md next — it lists the 9 tasks still needing harnesses, repo reuse hints, ordering, and per-task done-criteria.

Files

What ships with it

10 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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.

2mo ago First seen · 21 lines · 79 tokens per session scan A b35601b2f2ac

Subscribe to this mod's changes

harness-gen is a skill published in the GitHub repository specula-org/SysMoBench (26 stars, last pushed 18d ago), licensed Apache-2.0. It adds 79 tokens to every session and 353 once invoked, about $0.0003 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 2026-08-30.

Testing

research-engineer

An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.

davila7/claude-code-templates · 43 tokens

tika-eval-compare

Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".

apache/tika · 50 tokens

cua-driver

Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.

davidondrej/skills · 68 tokens

harness-gen