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 agentmods add instructions/thotischner/observability-mcp/claude-mdgit clone --depth 1 https://github.com/ThoTischner/observability-mcpWrote 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/instructions/thotischner/observability-mcp/claude-md)<a href="https://agentmods.dev/instructions/thotischner/observability-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/thotischner/observability-mcp/claude-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.01961 | $0.01961 |
| Opus 5 | $0.00981 | $0.00981 |
| Sonnet 5 | $0.00392 | $0.00392 |
| Haiku 4.5 | $0.00196 | $0.00196 |
Grade A, and why
observability-mcp CLAUDE.md 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 4d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -X POST http://localhost:8081/chaos/high-cpu How it starts
The opening of the file, as written. The whole thing — 157 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Observability-MCP Development Guide
Docker-First Development
This project is fully containerized. Never run npm install on the host machine.
Quick Start
docker compose --profile demo up --build
The demo stack brings up:
- a single-node k3s cluster (
rancher/k3sin a privileged container) - the three chaos-able example services (
api-gateway,payment-service,order-service) as Kubernetes Deployments inside k3s (namespaceomcp-demo) - Prometheus, Loki, Promtail (on the docker-compose side) scraping k3s NodePorts and pod logs
- the MCP server with the
kubernetestopology source pre-wired against k3s - the autonomous agent
The same Deployments that emit metrics and logs are what shows up in the
topology graph — that is what lets the agent correlate a metric/log
anomaly with the underlying host via get_blast_radius.
Without --profile demo, only mcp-server runs — for production deployments where Prometheus/Loki are managed elsewhere.
Rebuild a Single Service
docker-compose build mcp-server
docker-compose up -d mcp-server
View Logs
docker compose logs -f agent # Agent detection loop
docker compose logs -f mcp-server # MCP server
# Example services run inside k3s now — use kubectl to follow their logs:
docker exec observability-mcp-k3s-1 \
kubectl -n omcp-demo logs -f deployment/payment-service
Run Unit Tests
docker run --rm -w /app -v "$(pwd)/mcp-server:/app" node:20-alpine \
sh -c "npm install --silent && npx tsx --test src/analysis/*.test.ts"
Trigger Chaos (Demo)
curl -X POST http://localhost:8081/chaos/high-cpu
curl -X POST http://localhost:8081/chaos/error-spike
curl -X POST http://localhost:8081/chaos/slow-responses
curl -X POST http://localhost:8081/chaos/memory-leak
curl -X POST http://localhost:8081/chaos/reset
Chaos modes are correlated: error-spike also increases CPU + latency + error logs. Memory-leak generates OOM warnings.
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.
- 4d ago First seen · 157 lines · 1,961 tokens per session scan A 01212d9d5109
observability-mcp CLAUDE.md is an instructions file published in the GitHub repository ThoTischner/observability-mcp (6 stars, last pushed 4d ago), licensed Apache-2.0. It adds 1,961 tokens to every session, about $0.0098 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
charts CLAUDE.md
Claude Code instructions for duyet/charts, covering claude.md - project philosophy & development guide, the vision, core principles, 1. simplicity without losing power and 2. consistency is kindness.
helm-zabbix CLAUDE.md
Instructions for zabbix-community/helm-zabbix, covering claude.md, repository overview, common commands, architecture and template/value organization.
openclaw-kubernetes AGENTS.md
Instructions for feiskyer/openclaw-kubernetes, covering agents.md, links, development commands, lint the chart against all values files and render templates to verify correctness (output discarded).
openclaw-kubernetes CLAUDE.md
Instructions for feiskyer/openclaw-kubernetes, covering claude.md, links, development commands, lint the chart against all values files and render templates to verify correctness (output discarded).
appwrite AGENTS.md
AGENTS.md instructions for appwrite/appwrite, covering appwrite, commands, stack, layout and libraries.
plano CLAUDE.md
Instructions for katanemo/plano, covering claude.md, build & test commands, rust — wasm plugins (must target wasm32-wasip1), rust — brightstaff binary (native target) and rust — tests, format, lint.