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 skills add KaydenClark/LLM_Workbench --skill improve-harnessgit clone --depth 1 https://github.com/KaydenClark/LLM_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/skills/kaydenclark/llm_workbench/improve-harness)<a href="https://agentmods.dev/skills/kaydenclark/llm_workbench/improve-harness"><img src="https://agentmods.dev/badge/skills/kaydenclark/llm_workbench/improve-harness/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.
<a href="https://agentmods.dev/skills/kaydenclark/llm_workbench/improve-harness"><img src="https://agentmods.dev/badge/skills/kaydenclark/llm_workbench/improve-harness.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00086 | $0.02643 |
| Opus 5.5 | $0.00034 | $0.01057 |
| Sonnet 5.5 | $0.00017 | $0.00529 |
| Haiku 4.5 | $0.00009 | $0.00264 |
Grade A, and why
improve-harness 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 12d 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 — 260 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Improve Harness
Harness engineering improves what agents produce by shaping the environment around a fixed model and agent: the context it can reach, the tools it can call, the boundaries it works inside and the feedback that proves its work. This skill is the one procedure for improving that environment around one observed job. It runs one loop:
baseline -> earliest gap -> smallest owning intervention -> native verification -> fresh rerun -> retain, revise or remove
The loop's shape follows Ryan Lopopolo's improve-harness playbook (CC BY 4.0, linked and attributed on the LLM Workbench Wiki's harness engineering lineage page); this skill is written in the Workbench's own words and copies nothing. It replaces a family of review stages with one pass that either closes the job under the observed conditions or names the next missing boundary. Comparative, causal or longitudinal claims need the room's evaluation procedures, not this loop.
Purpose
Take one bounded, representative job that a fixed worker ran in this room, find the earliest point where the harness failed it, make the smallest change at the owner that can shape future runs, prove the change through the room's own checks and a fresh rerun, and then decide whether the change earns its carrying cost. The durable result is a better environment and its evidence, not another harness component.
Method
Read before changing. Every step until the fourth is read-only: the baseline
and the gap are established from the room's controls, source, tests, records
and the job's trajectory with non-mutating checks only, and a change to
Actuality happens only inside the authority the current request and
AGENTS.md grant. Read each claim by the Governance Plane it plays in this
operation: Canon (the binding current-state rule), Grounding (evidence about
whether the work was done correctly), Actuality (the target as it is) and
Enduring Context (durable reference). A control that says a thing is not
proof the thing happened, and a passing check is not proof the worker was
expected to attach it.
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.
- 12d ago First seen · 260 lines · 86 tokens per session scan A 39828d70ef23
improve-harness is a skill published in the GitHub repository KaydenClark/LLM_Workbench (2 stars, last pushed 2d ago), licensed MIT. It adds 86 tokens to every session and 2,643 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 2026-10-07.
Other skills, from other repositories
debug-optimize-lcp
Guides debugging and optimizing Largest Contentful Paint (LCP) using Chrome DevTools MCP tools. Use this skill whenever the user asks about LCP performance, slow page loads, Core Web Vitals optimization, or wants to understand why their page's main content takes too long to appear. Also use when the user mentions…
systematic-debugging
Use when debugging a failing test, build error, or runtime issue that isn't immediately obvious. Guides a 4-phase root cause analysis instead of random fix attempts.
log-error-digest
Analyze log files to troubleshoot errors, identify peak error periods, and produce error clustering, frequency statistics, and time distribution reports. Supports JSON, syslog, and Nginx formats with automatic detection. Use when a user uploads a .log file and asks to analyze errors, find patterns, debug issues, or…
deck-retro
Run a fully local agent-deck retrospective over the user's own transcripts, Recall index and logs. Use when a user wants to find recurring failures or take their own finding from investigation through synthetic reproduction, a user-filed issue, a test-first fix and contributor PR handoff. Also use for weekly usage…
incident-response-incident-response
Coordinate production incident triage, mitigation, investigation, recovery, and follow-up. Use when responding to outages or other incidents that require structured operational coordination.
ci-fixer
CI failures - read error, minimal fix, verify.