Agent Observability

Agent Observability is an AI Engineering skill for Claude Code, published by mohitagw15856 in pm-claude-skills.

1.3K stars231 forkson mohitagw15856/pm-claude-skillsAdded 2026/08/20Repository updated 2026/08/20

agent-skillsagentsai-agentsai-toolsanthropicchatgptclaudeclaude-codeclaude-skillscursordeveloper-toolsgeminillmmcpmcp-serverproduct-managementproductivityprompt-engineeringpromptsskills

Install in seconds

Install Agent Observability

Copy Agent Observability into your Claude Code skills folder. Run the command in your terminal, or review the source on GitHub before installing.

terminal

npx degit https://github.com/mohitagw15856/pm-claude-skills/tree/main/skills/agent-observability-spec ~/.claude/skills/agent-observability-spec

Requires Node.js. Downloads this skill only — not the rest of the repository — into your Claude Code skills folder.

Without Node.js

git clone https://github.com/mohitagw15856/pm-claude-skills.git

Clones the whole repository, then copy the skill’s own directory into your skills folder yourself.

In this catalog

Source file: skills/agent-observability-spec/SKILL.md in mohitagw15856/pm-claude-skills
Installs to: ~/.claude/skills/agent-observability-spec
Collection: One of 25 skills cataloged from this repository
Category: AI Engineering — 2451 skills

What Agent Observability does

Agent Observability specifies the tracing, metrics, and alerts for an AI agent or LLM feature in production. Use it when you need to log LLM app behavior, design agent spans, define quality and cost monitors, or determine if an agent is misbehaving.

Agent Observability is cataloged under AI Engineering on DirSkills. Agent Observability comes from a repository tagged agent-skills, agents, ai-agents, ai-tools and anthropic.

Documentation

README

Excerpt

Agent Observability Spec Skill

You can't fix what you didn't record. For LLM systems the unit of observability is the trace — everything the model saw and did — because behaviour, not uptime, is what fails. This skill specifies what to capture, what to compute from it, and when to page someone.

What This Skill Produces

A trace schema: per-request spans and the fields each must carry
Metric definitions across health, quality, cost, and behaviour — each with a threshold and owner
A sampling and retention policy that keeps cost sane and debugging possible
A privacy note: what logged content contains, who can see it, and how long it lives

This is the opening of the README. Read the full README on GitHub.

Frequently asked about Agent Observability

What else does mohitagw15856 publish alongside Agent Observability?
Agent Observability is one of 25 skills that DirSkills catalogs from mohitagw15856/pm-claude-skills, the repository it ships in. Its siblings there include 360 Feedback Template, 401k Plan Decoder and A/B Test Planner. Each one is a separate skill with its own page in this directory, installs the same way Agent Observability does, and is maintained by mohitagw15856 in that same repository. The rest of the collection is listed on the mohitagw15856/pm-claude-skills page.
How does Agent Observability compare to other AI Engineering skills?
Agent Observability ranks #1115 by stars among the 2451 AI Engineering skills in this catalog. The most-starred ones next to it are Architecture Decision Records, AI-First Engineering and Agentic OS. DirSkills ranks by the star count of the repository each skill ships in, so that order reflects how popular those repositories are rather than any review of Agent Observability against them. Open each page to compare what they document and how they install.