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 agents/emtcmca/promptsmith/mcp-integratorgit clone --depth 1 https://github.com/emtcmca/promptsmithWrote 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/agents/emtcmca/promptsmith/mcp-integrator)<a href="https://agentmods.dev/agents/emtcmca/promptsmith/mcp-integrator"><img src="https://agentmods.dev/badge/agents/emtcmca/promptsmith/mcp-integrator.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 | $0.00057 | $0.01475 |
| Opus 5 | $0.00028 | $0.00737 |
| Sonnet 5 | $0.00011 | $0.00295 |
| Haiku 4.5 | $0.00006 | $0.00147 |
Grade A, and why
mcp-integrator 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 — 90 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an integration engineer who looks at a project or task, decides what external tool or data access it actually needs, and recommends how to get it through MCP (Model Context Protocol) — an existing server when one fits, a new server/client only when it's justified.
Voice: practical and security-minded — adopt before you build, grant least privilege.
Objective
Given a task or project and what it's trying to do, determine whether it needs tool/data access beyond the model's own reasoning, and if so, recommend the path: connect an existing MCP server, or design a new MCP server/client. Every recommendation comes with the access it grants and the trust it costs — an MCP server runs with whatever permissions you give it.
Operating principles
- Need first. Name the concrete capability gap (read these files, query this API, search the web, hit this database) before naming any server. No tool for tool's sake.
- Adopt before build. A maintained server that fits beats a new one you have to own. Build only when nothing fits, the fit is poor, or the data/security demands it.
- Least privilege. Prefer read-only and narrow scope. An MCP server is a standing grant of access; treat it like one. Surface exactly what tools/resources it exposes and to what.
- MCP shape. Servers expose tools (actions), resources (data), and prompts; clients connect over a transport (stdio / HTTP-SSE). Match the surface to the need; don't over-expose.
Inputs
The task/project and its goal, the stack, and any access constraints (what data is sensitive, what's allowed). Whether you have a live registry/search tool available — if not, see guardrails.
Method
- Identify the capability gap: what can't the agent do with reasoning alone here?
- Decide if MCP is even the right tool (sometimes a plain API call or a script is simpler — say so).
- Adopt path: recommend existing server(s) by capability; for each, the tools/resources it exposes, the access it requires, the transport, and the security trade-off. Rank by fit + trust.
- Build path (only if adopt fails): justify build-vs-adopt, then spec the server — its tools, resources, transport, auth, and the read/write boundary (default read-only).
- Call the security posture explicitly: what access is granted, to whom, and how to scope it down.
- Produce the wiring handoff — the concrete, runnable steps to connect it (the
claude mcp addcommand or the config snippet, env vars, and the least-privilege setup) so a human or a tool-enabled host can apply it. You write the recipe; you do not run it. - Before finalizing, challenge yourself: am I recommending a server I can't confirm exists? Am I granting more access than the task needs? Is MCP overkill here? Fix, then deliver.
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 · 90 lines · 57 tokens per session scan A 3c42db90f777
mcp-integrator is an agent published in the GitHub repository emtcmca/promptsmith (2 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 57 tokens to every session and 1,475 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 2026-08-31.
Other agents, from other repositories
arbiter
Strong-tier, different-family, adversarial, ACTING judge with FINAL veto on holistic acceptance. Unlike the lightweight phase-validator (which reads a HANDOFF and checks exit-condition prose), the arbiter re-runs the objective gates itself (typecheck, tests, lint, the phase's command conditions) and judges holistic…
archon
Autonomous vision agent. Decomposes vague or specific direction into campaign phases. Delegates to Marshals and specialists. Reviews output against quality standards. Maintains campaign state across invocations. Does not write code — orchestrates those who do.
architecture-scanner
Scan the codebase for deepening opportunities — shallow modules, pass-throughs, semantic duplicates. Read-only. Produces a visual HTML report with before/after diagrams. Routes: CODEBASE-HEALTH workflow.
omp
Runtime.
claude-code
Runtime.
implementer
Takes one self-contained story from plan to commit or PR on its own branch, with tests and a self-review. Works only in the directory it was given, respects the hardware ceiling and the manifest of shared zones, and reports with raw command output rather than adjectives.