observability-llm-obs

observability-llm-obs is a skill for Claude Code, Codex from aspectrr/deer. It costs 52 tokens per session (887 once invoked), scanned A, original, MIT.

An observability guide for monitoring large language models and agent-based applications through data stored in Elastic.

In plain words
What is it for?
Use it to inspect trace, metric, and log data for model providers, model names, token counts, latency, errors, outcomes, and parent-child call relationships.
Why use it?
It helps identify slow or failed calls, token and cost usage, response quality signals, and relationships between chained model calls.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/aspectrr/deer/observability-llm-obs
Any agent
npx skills add aspectrr/deer --skill observability-llm-obs
Clone the repo
git clone --depth 1 https://github.com/aspectrr/deer

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for observability-llm-obs

README.md
[![agentmods](https://agentmods.dev/badge/skills/aspectrr/deer/observability-llm-obs.svg)](https://agentmods.dev/skills/aspectrr/deer/observability-llm-obs)
Your own site
<a href="https://agentmods.dev/skills/aspectrr/deer/observability-llm-obs"><img src="https://agentmods.dev/badge/skills/aspectrr/deer/observability-llm-obs.svg" alt="Measured on agentmods" height="20"></a>
Per session 52 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 887 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00052 $0.00887
Opus 5 $0.00026 $0.00443
Sonnet 5 $0.00010 $0.00177
Haiku 4.5 $0.00005 $0.00089

Measured 5d ago against content hash 5a35fa284182, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

observability-llm-obs 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 5d 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.

deer-cli/internal/skill/defaults/observability-llm-obs/SKILL.md · 94 lines

How it starts

The opening of the file, as written. The whole thing — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.

LLM and Agentic Observability

Monitor LLMs and agentic components using data ingested into Elastic. Focus on performance, cost/token utilization, response quality, and call chaining.

Where to look

  • Trace data (APM / OTel): traces* for LLM spans from OTel/EDOT instrumentations
  • Integration metrics/logs: metrics* and logs* from Elastic LLM integrations (OpenAI, Azure, Bedrock, Vertex AI)
  • Discover first: Use GET _data_stream or GET traces*/_mapping to find available data

Data available

From traces (traces*)

Purpose Example attribute names (OTel GenAI)
Operation / provider gen_ai.operation.name, gen_ai.provider.name
Model gen_ai.request.model, gen_ai.response.model
Token usage gen_ai.usage.input_tokens, gen_ai.usage.output_tokens
Errors error.type

Use duration and event.outcome for latency and success/failure. Use trace.id and parent/child relationships for call chaining analysis.

Use cases and query patterns

LLM performance

FROM traces*
| WHERE @timestamp >= "2025-03-01T00:00:00Z" AND @timestamp <= "2025-03-01T23:59:59Z"
  AND span.attributes.gen_ai.provider.name IS NOT NULL
| STATS request_count = COUNT(*), failures = COUNT(*) WHERE event.outcome == "failure",
    avg_duration_us = AVG(span.duration.us)
  BY span.attributes.gen_ai.request.model
| EVAL error_rate = failures / request_count
| LIMIT 100

Token usage over time

FROM traces*
| WHERE @timestamp >= "2025-03-01T00:00:00Z" AND @timestamp <= "2025-03-01T23:59:59Z"
  AND span.attributes.gen_ai.provider.name IS NOT NULL
| STATS input_tokens = SUM(span.attributes.gen_ai.usage.input_tokens),
    output_tokens = SUM(span.attributes.gen_ai.usage.output_tokens)
  BY BUCKET(@timestamp, 1 hour), span.attributes.gen_ai.request.model
| SORT @timestamp
| LIMIT 500

Read the full file on GitHub · 94 lines

Changes

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.

  1. 5d ago First seen · 94 lines · 52 tokens per session scan A 5a35fa284182

Subscribe to this mod's changes

observability-llm-obs is a skill published in the GitHub repository aspectrr/deer (404 stars, last pushed 4mo ago), licensed MIT. It adds 52 tokens to every session and 887 once invoked, about $0.0003 per session on Opus 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 2026-08-30.