SkillsBench is a benchmark for measuring how effectively AI agents use modular skills—folders containing instructions, scripts, and resources—to complete specialized tasks. It helps researchers and developers evaluate both skill quality and agent behavior, including tasks that require combining multiple skills. The catalogue’s skills and instructions are evaluated as part of this workflow.
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 benchflow-ai/skillsbench --skill glm-calibrationgit clone --depth 1 https://github.com/benchflow-ai/skillsbenchWrote 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/benchflow-ai/skillsbench/glm-calibration)<a href="https://agentmods.dev/skills/benchflow-ai/skillsbench/glm-calibration"><img src="https://agentmods.dev/badge/skills/benchflow-ai/skillsbench/glm-calibration/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/benchflow-ai/skillsbench/glm-calibration"><img src="https://agentmods.dev/badge/skills/benchflow-ai/skillsbench/glm-calibration.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00032 | $0.00839 |
| Opus 5 | $0.00016 | $0.00419 |
| Sonnet 5 | $0.00006 | $0.00168 |
| Haiku 4.5 | $0.00003 | $0.00084 |
Grade A, and why
glm-calibration scanned grade A with 1 finding 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 8d 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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
subprocess.run(['glm'], capture_output=True) Copies of this mod
1 near-identical copy found in the catalogue:
- glm-calibration — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GLM Calibration Guide
Overview
GLM calibration involves adjusting physical parameters to minimize the difference between simulated and observed water temperatures. The goal is typically to achieve RMSE < 2.0°C.
Key Calibration Parameters
| Parameter | Section | Description | Default | Range |
|---|---|---|---|---|
Kw |
&light |
Light extinction coefficient (m⁻¹) | 0.3 | 0.1 - 0.5 |
coef_mix_hyp |
&mixing |
Hypolimnetic mixing coefficient | 0.5 | 0.3 - 0.7 |
wind_factor |
&meteorology |
Wind speed scaling factor | 1.0 | 0.7 - 1.3 |
lw_factor |
&meteorology |
Longwave radiation scaling | 1.0 | 0.7 - 1.3 |
ch |
&meteorology |
Sensible heat transfer coefficient | 0.0013 | 0.0005 - 0.002 |
Parameter Effects
| Parameter | Increase Effect | Decrease Effect |
|---|---|---|
Kw |
Less light penetration, cooler deep water | More light penetration, warmer deep water |
coef_mix_hyp |
More deep mixing, weaker stratification | Less mixing, stronger stratification |
wind_factor |
More surface mixing | Less surface mixing |
lw_factor |
More heat input | Less heat input |
ch |
More sensible heat exchange | Less heat exchange |
Calibration with Optimization
from scipy.optimize import minimize
def objective(x):
Kw, coef_mix_hyp, wind_factor, lw_factor, ch = x
# Modify parameters
params = {
'Kw': round(Kw, 4),
'coef_mix_hyp': round(coef_mix_hyp, 4),
'wind_factor': round(wind_factor, 4),
'lw_factor': round(lw_factor, 4),
'ch': round(ch, 6)
}
modify_nml('glm3.nml', params)
# Run GLM
subprocess.run(['glm'], capture_output=True)
# Calculate RMSE
rmse = calculate_rmse(sim_df, obs_df)
return rmse
# Initial values (defaults)
x0 = [0.3, 0.5, 1.0, 1.0, 0.0013]
# Run optimization
result = minimize(
objective,
x0,
method='Nelder-Mead',
options={'maxiter': 150}
)
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.
- 8d ago First seen · 93 lines · 32 tokens per session scan A a1342ccb6297
glm-calibration is a skill published in the GitHub repository benchflow-ai/skillsbench (1,764 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 32 tokens to every session and 839 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
run-aeon-benchmark
Use when asked to run, benchmark, evaluate, or score an LLM with AEON Bench. You run the AEON Bench Pod on the user's machine, point it at a model, run the benchmark, and submit the signed result to the public leaderboard at aeon-bench.com. All work happens on the pod. The mothership only shows the board and accepts…
data-matching
Matching observation data to simulation output with exact datetime and depth binning.
model-calibration
Calculating RMSE metrics and parameter calibration for lake model validation.
dbscan-custom-metric
Run DBSCAN clustering with a custom distance metric using sklearn, including how to define weighted Euclidean metrics and extract cluster centroids.
greedy-bipartite-matching
Implement greedy bipartite matching for pairing cluster centroids with expert annotations.
plate-tectonics-geospatial
Analyze plate tectonics data using GeoPandas, identify points within plates, and calculate distances to boundaries.