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/jperelli/graylog-mcp/claude-mdgit clone --depth 1 https://github.com/jperelli/graylog-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/jperelli/graylog-mcp/claude-md)<a href="https://agentmods.dev/instructions/jperelli/graylog-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/jperelli/graylog-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 | $0.01946 | $0.01946 |
| Opus 5 | $0.00973 | $0.00973 |
| Sonnet 5 | $0.00389 | $0.00389 |
| Haiku 4.5 | $0.00195 | $0.00195 |
Grade A, and why
graylog-mcp CLAUDE.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 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.
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 — 115 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
What this is
A minimal MCP (Model Context Protocol) server that lets an AI agent query one or
more Graylog instances. Published to npm as @jperelli/graylog-mcp and run via
npx. Pure ESM Node.js, no build step.
Layout
src/index.js— the entire server. Single file: instance loading, tool schemas, and handlers all live here.package.json—binmapsgraylog-mcp→src/index.js(has a#!/usr/bin/env nodeshebang).README.md— user-facing setup, tool reference, and design rationale.example-config.json— sample MCP client config.test/—npm test(node --test test/*.test.js). Dependency-free smoke tests, one file per tool (list_streams,search,analyze,get_message).test/helpers.jsstands up a mock Graylog overnode:httpand launchessrc/index.jsover stdio via the MCP client; each*.test.jsfile drives one tool against it. No live Graylog needed.
Run / debug
GRAYLOG_BASE_URL_INSTANCE_1=http://host:9000 \
GRAYLOG_API_TOKEN_INSTANCE_1=<token> \
node src/index.js # speaks MCP over stdio
Set DEBUG=true for verbose logging to stderr. The transport is stdio, so anything on
stdout that isn't a JSON-RPC frame breaks the protocol — log to stderr only
(console.error).
Configuration model
Instances are configured via numbered env vars: GRAYLOG_BASE_URL_INSTANCE_N,
GRAYLOG_API_TOKEN_INSTANCE_N, and optional GRAYLOG_LABEL_INSTANCE_N. Instance 1 also
falls back to the legacy BASE_URL / API_TOKEN. Only instances with both URL and token
are active; the first is the default.
Tools (all in src/index.js)
search— read matching log lines; concise projection by default,verbose/fieldsfor full, absolutefrom/tosupported. On 0 matches it self-diagnoses (see below).analyze— top values of a field + optional time histogram;valueContainssubstring-matches the values.list_fields— which fields actually exist, filtered bycontains.list_streams— readable stream IDs. Rarely needed; capped output.get_message— fetch one full document by_id+_index.
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 · 115 lines · 1,946 tokens per session scan A 6f0b6990dff0
graylog-mcp CLAUDE.md is an instructions file published in the GitHub repository jperelli/graylog-mcp (0 stars, last pushed 1mo ago), licensed MIT. It adds 1,946 tokens to every session, about $0.0097 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 2026-08-31.
Other instructions, from other repositories
mcp-deep-search CLAUDE.md
Claude Code instructions for just-every/mcp-deep-search, covering claude.md, project overview, core modules & files, commands and development.
memoryClaw CLAUDE.md
Instructions for tostechbr/memoryClaw, covering memoryclaw, project status, ✅ resolved: multi-user isolation (sprint 0 — 2026-03-06), next steps (sprint 2 focus) and sprint 0: multi-user isolation ✅ completa (2026-03-06).
mish CLAUDE.md
Instructions for aetherwing-io/mish, covering mish, status, naming, docs and three modes, one binary.
ultimate_mcp_server AGENTS.md
AGENTS.md instructions for Dicklesworthstone/ultimate_mcp_server, covering agents.md - ultimate mcp server, rule 1 - no file deletion, irreversible git & filesystem actions, project overview and key directories.
mish AGENTS.md
Instructions for aetherwing-io/mish, covering agent instructions, quick reference and landing the plane (session completion).
hoop CLAUDE.md
Claude Code instructions for hoophq/hoop, covering claude.md, project overview, toolchain & prerequisites, architecture and module breakdown.