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/muchiny/bridge-mcp/claude-mdgit clone --depth 1 https://github.com/muchiny/bridge-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/muchiny/bridge-mcp/claude-md)<a href="https://agentmods.dev/instructions/muchiny/bridge-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/muchiny/bridge-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 | $0.04830 | $0.04830 |
| Opus 5 | $0.02415 | $0.02415 |
| Sonnet 5 | $0.00966 | $0.00966 |
| Haiku 4.5 | $0.00483 | $0.00483 |
Grade A, and why
bridge-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 — 315 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Project Overview
Bridge MCP (binary bridge-mcp, formerly MCP SSH Bridge / mcp-ssh-bridge) is a Rust MCP server that enables Claude Code to securely execute commands on air-gapped environments via SSH. JSON-RPC over stdio, strict security controls. 476 tools across 77 groups (62 Linux, 13 Windows, 2 cross-platform). Count source of truth: .migration-baseline.json (python3 scripts/validate_baseline.py).
CLI-as-Tool Mode (Alternative to MCP)
All 476 MCP tools are accessible directly via CLI, enabling 10-32x token savings compared to MCP mode. Use CLI for dev workflows, MCP for enterprise integration.
Quick Reference
# Invoke any tool directly (same as MCP, but via CLI)
bridge-mcp tool ssh_docker_ps host=prod
bridge-mcp tool ssh_exec host=prod command="df -h" --json
bridge-mcp tool ssh_k8s_get --json-args '{"host":"k8s","resource":"pods","namespace":"default"}'
# Progressive discovery (token-efficient for AI agents)
bridge-mcp list-tools --groups-only # 77 groups (~2K tokens)
bridge-mcp list-tools --group docker # tools in group (~500 tokens)
bridge-mcp list-tools --search kubernetes # keyword search
bridge-mcp describe-tool ssh_docker_ps # full schema + Reduction Strategy (~200 tokens)
bridge-mcp describe-tool ssh_exec --json # schema as JSON
# Global --json flag works on all commands
bridge-mcp --json status
bridge-mcp --json tool ssh_service_status host=web1 service=nginx
Token-efficient patterns (IMPORTANT for AI agents)
Always call describe-tool before invoking an unknown tool — its top-of-output
Reduction Strategy line tells you which params apply. Server-side filtering
runs BEFORE truncation, so you never lose data to the output cap.
| Output kind | Strategy | Example |
|---|---|---|
Tabular (docker_ps, service_list) |
columns + limit |
columns='["NAME","STATUS"]' limit=20 |
Json (k8s_get, docker_inspect, awx_*) |
jq_filter + output_format=tsv |
jq_filter='.items[] | [.name, .status]' output_format=tsv (60-80% savings) |
| Yaml | yq_filter + output_format=tsv |
same shape as jq |
| Auto | Any of the above | tool auto-detects |
RawText (logs, ssh_exec) |
save_output=/tmp/out.txt |
read file locally afterwards |
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 · 315 lines · 4,830 tokens per session scan A 1c92685f4c35
bridge-mcp CLAUDE.md is an instructions file published in the GitHub repository muchiny/bridge-mcp (8 stars, last pushed 4d ago), licensed MIT. It adds 4,830 tokens to every session, about $0.0242 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
epa-mcp-server CLAUDE.md
Claude Code instructions for cyanheads/epa-mcp-server, covering developer protocol, what's next?, core rules, patterns and tool.
ai-workflow AGENTS.md
Instructions for cunhaax/ai-workflow, covering ai workflow template, rules — non-negotiable, project overview, commands and architecture.
trackly-cli CLAUDE.md
Claude Code instructions for trackly-app/trackly-cli, covering trackly-cli, tech stack, backend production source of truth, directory structure and key commands.
vibetags com-example-config-AsyncTestConfig.instructions.md
Instructions for PIsberg/vibetags, covering copilot instructions for asynctestconfig and immutable type.
McpServer AGENTS.md
AGENTS.md instructions for sharpninja/McpServer, covering agent instructions, session start, rules, byrd test gate and where things live.
AmbyKit CLAUDE.md
Instructions for ambystechcom/AmbyKit, covering claude code — notes for the ambykit repo and claude-specific.