Agent skills for Claude Desktop

Automate Beeminder tasks via Rube MCP (Composio).

Beeper cache: contact hints, room lookup, WhatsApp/iMessage traces, FTS.

Enforce 2026 folder structure standards - feature-based organization, max nesting depth, unidirectional imports.

Captures before/after screenshots of web pages or elements for visual comparison.

Fire the moment the user proposes a build.

Executes elements whose implementation is a python:// entry point, with the study's dependencies installed.

Behavioral design validation for game development.

X-ray any AI model's behavioral patterns — refusal boundaries, hallucination tendencies, reasoning style, formatting defaults.

X-ray any AI model's behavioral patterns — refusal boundaries, hallucination tendencies, reasoning style, formatting defaults.

Go behind a program to discover what it ACTUALLY uses — its real dependencies, libraries, syscalls, files, network endpoints, env/config…

../../../engineering/behuman/skills/behuman/SKILL.md

Scientific data analysis including data cleaning, exploratory data analysis (EDA), statistical testing, regression, and reporting.

Iterative daily-newsletter writing workflow for Ben AI's audience.

Automate Bench tasks via Rube MCP (Composio).

Run mxx repository programs, tests, or benchmarks on Runpod instances with GPU machine selection, network volume setup, SSH/scp…

Read artifacts from the shared bench — the workspace where desks leave findings, verdicts, and work products for each other and the…

Benchmark a model served on MAX with the `max benchmark` command: measure throughput (tokens/sec), latency (TTFT, TPOT, inter-token…

Design and calibrate a multi-Case capability Benchmark and establish a traceable Formal Baseline.

Automate Benchmark Email tasks via Rube MCP (Composio).

Use after competitive-platform-analysis has produced a tiered competitor set.

Find, inspect, and check AI benchmark records with the Benchmark Radar CLI.