Akash Pal
Community@akashjpal
Akash Pal's registry delivers Arize AX skills for LLM observability, evaluation, prompt optimization, and regulatory compliance auditing.
Agent Skills by Akash Pal
Showing 16 vetted skills indexed across 1 GitHub repositories.
akashjpalarize-trace
Export and inspect Arize traces and spans to debug LLM application behavior.
akashjpalarize-experiment
Creates, runs, and compares Arize experiments for evaluating model performance via the ax CLI.
akashjpalarize-link
Generates deep links to Arize UI traces, spans, sessions, datasets, and evaluators.
akashjpalfind-skills
Searches and installs agent skills from the open skills ecosystem using the Skills CLI.
akashjpalarize-compliance-audit
Audits AI agent codebases for regulatory compliance gaps across EU, US, and ISO frameworks.
akashjpalarize-instrumentation
Adds Arize AX tracing to LLM applications via a two-phase analyze-then-implement workflow.
akashjpalbrainstorming
Guides collaborative design exploration and spec writing before any implementation work begins.
akashjpalarize-dataset
Creates, manages, and queries Arize datasets and examples using the ax CLI.
akashjpalarize-ai-provider-integration
Manages Arize AI integrations storing LLM provider credentials via the ax CLI.
akashjpalarize-annotation
Creates annotation configs and queues and applies human labels to Arize spans.
akashjpaltdd
Guides test-driven development using the red-green-refactor loop with seam-based testing.
akashjpalgrill-me
Runs a structured interview session to challenge and refine a plan or design.
akashjpalgrilling
Interviews users in structured rounds to stress-test plans and decisions.
akashjpalarize-evaluator
Creates and runs LLM-as-judge evaluators on Arize via the ax CLI.
akashjpalarize-prompt-optimization
Optimizes LLM prompts using Arize trace data, evaluations, and a data-driven iteration loop.
akashjpalfix-bug
Diagnose bugs through a gated RCA, brainstorm, implement, and walkthrough workflow.
Frequently Asked Questions About Akash Pal
FAQPage SchemaWhat tasks can I perform with Akash Pal's Arize skills?▼
You can export and inspect traces and spans, create and compare model experiments, manage datasets and golden test sets, run LLM-as-judge evaluations, optimize prompts from production trace data, configure annotation queues, manage LLM provider credentials, and audit codebases for regulatory compliance.
Who are these skills designed for?▼
They target ML engineers, LLM application developers, and AI platform teams who need observability, evaluation, and compliance coverage for production LLM systems. Secondary skills like tdd, fix-bug, and brainstorming support general software engineers practicing test-first development and structured debugging.
What are the prerequisites for running the Arize skills?▼
Most Arize skills require the ax CLI and a configured Arize profile. The evaluator skill additionally needs an AI integration storing LLM provider credentials. Instrumentation supports Python and TypeScript/JavaScript via openinference packages, and Java or Go via the OpenTelemetry SDK with manual OpenInference spans.
How do I add tracing to an existing LLM application?▼
Use the arize-instrumentation skill, which follows a two-phase flow: it first analyzes your codebase, then implements Arize AX tracing after your confirmation. Python and TypeScript apps get auto-instrumentation through openinference packages, while Java and Go require manual OpenInference spans via OpenTelemetry.
Which compliance frameworks does the audit skill cover?▼
The arize-compliance-audit skill covers the EU AI Act, GPAI Code of Practice, GDPR, NIST AI RMF, Colorado AI Act, HIPAA, and ISO 42001. It scans your codebase for gaps, cross-references Arize instrumentation for audit trail coverage, and outputs a framework-specific remediation checklist.