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/themoah/klag/agents-mdgit clone --depth 1 https://github.com/themoah/klagWrote 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/themoah/klag/agents-md)<a href="https://agentmods.dev/instructions/themoah/klag/agents-md"><img src="https://agentmods.dev/badge/instructions/themoah/klag/agents-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 | $0.00962 | $0.00962 |
| Opus 5 | $0.00481 | $0.00481 |
| Sonnet 5 | $0.00192 | $0.00192 |
| Haiku 4.5 | $0.00096 | $0.00096 |
Grade A, and why
klag AGENTS.md 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 yesterday.
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.
How it starts
The opening of the file, as written. The whole thing — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Instructions for AI coding agents working in this repository. CLAUDE.md is the detailed
reference — architecture, every environment variable, every metric. Read it before changing
code; this file is the short version plus the rules that are easy to violate.
What this is
Klag — a Kafka consumer lag exporter. Java 21 on Vert.x 4.5.30, Micrometer for metrics, Apache-2.0. It polls Kafka's AdminClient read-only and exports lag and derived signals to Prometheus, Datadog, or OTLP. Docs: https://klag.dev (machine-readable at https://klag.dev/llms.txt).
Are you trying to use Klag rather than develop it — install it, connect it, read its
metrics? Then you want the skill, not this repo: /plugin marketplace add themoah/klag,
or fetch https://klag.dev/.well-known/agent-skills/index.json. There is also a read-only
documentation MCP server at https://klag.dev/mcp.
Build and test
Java 21 required (sdk use java 21.0.9-tem). Use ./gradlew locally; CI calls gradle
directly, as no wrapper JAR is committed.
./gradlew compileJava
./gradlew test
./gradlew assemble # fat JAR
./scripts/test-helm-chart.sh # chart templating, offline
./scripts/check-plugin.sh # plugin manifest shape
cd website && npm run build && npm test # docs site + generators
./scripts/e2e-test.sh and ./scripts/e2e-strimzi-test.sh spin up real clusters — slow,
and they need Docker. Do not run them speculatively.
Invariants — do not break these
- Admin request volume must stay independent of topic count. Each collection cycle
fetches committed offsets per group, then makes one batched
getLogEndOffsets(Set)call for the union of their topics. Reintroducing a per-topic fetch turns one call into four requests × every topic × every cycle.MetricsCollectorBatchingTestpins this. - Deleted topics are filtered before
describeTopics. One unknown topic fails the whole batch, and committed offsets outlive a deleted topic foroffsets.retention.minutes. A failedlistTopicsmust propagate, never fall through unfiltered. - Partial cycles skip stale-gauge cleanup but still publish the MCP snapshot. Cleaning against an incomplete key set would delete live series. An empty snapshot is never published.
- Adding a metric means updating
dashboard/demo-dashboard.jsonand the metrics tables inCLAUDE.mdandwebsite/src/content/docs/metrics/. - Version bumps. One bump of
versioninbuild.gradle.ktsper PR that changes the application, andcharts/klag/Chart.yamlappVersionplus the artifacthub annotation must match it. Website-only or docs-only changes do not bump.
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.
- yesterday First seen · 72 lines · 962 tokens per session scan A 278b75e34049
klag AGENTS.md is an instructions file published in the GitHub repository themoah/klag (82 stars, last pushed 4d ago), licensed Apache-2.0. It adds 962 tokens to every session, about $0.0048 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 2026-09-03.
Other instructions, from other repositories
kafka-backup CLAUDE.md
Claude Code instructions for osodevops/kafka-backup, covering project overview, key problems solved, repository structure, build & test commands and build the project.
RuView AGENTS.md
AGENTS.md instructions for ruvnet/RuView, covering ruview repository instructions for codex, operating contract, repository map, ruview contributor harness and homecore metaharness.
netdata AGENTS.md
AGENTS.md instructions for netdata/netdata, covering agents.md, goals, requirement language, working with the user and development principles.
worldmonitor AGENTS.md
AGENTS.md instructions for koala73/worldmonitor, covering agents.md, task mode and authority, start here, surface routing and architecture invariants.
nvidia_gpu_exporter AGENTS.md
AGENTS.md instructions for utkuozdemir/nvidia_gpu_exporter, covering nvidiagpuexporter — agent guide, what this repo is, the three collection backends, code map and the capture corpus and how testing works.
netdata CLAUDE.md
Claude Code instructions for netdata/netdata, a project described as: The fastest path to AI-powered full stack observability, even for lean teams.