# ai.promptarch/mcp

> Research-backed lint + generation for agent context files (CLAUDE.md, AGENTS.md) + prompt enhancer.

- name: ai.promptarch/mcp
- version: 1.2.0
- connection class: R0 (autonomous)
- trust: 82/100 flags: no-repository
- quality: 82/100 (good)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/ai.promptarch%2Fmcp
- tools: https://api.protogrid.dev/v1/servers/ai.promptarch%2Fmcp/tools

**An agent can connect right now, no human step.** Remote endpoint, no authentication, reachable on the last probe.

## Connection (mcpServers)

```json
{
  "mcpServers": {
    "promptarch": {
      "type": "http",
      "url": "https://mcp.promptarch.ai/mcp",
      "headers": {
        "Authorization": "Bearer ${PROMPTARCH_TOKEN}"
      }
    }
  }
}
```

Placeholders to fill: `${PROMPTARCH_TOKEN}`.

## Trust

- hygiene: 80
- liveness: 98
- freshness: 100
- provenance: 55
- drivers: +dns-namespace +website +reachable +uptime-96% +updated-4d-ago -no-repository
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 82/100 (good)

### protocol: 75 (weight 25)

- warn `protocol.modern`: newest supported version is 2025-11-25; 2026-07-28 not supported, older versions are deprecated until 2027-07-28
- n/a `protocol.stateless`: only applies to 2026-07-28 servers
- pass `protocol.transport`: streamable HTTP
- n/a `protocol.list_ttl`: only applies to 2026-07-28 servers

### authorization: n/a (weight 15)

- n/a `auth.prm`, `auth.as_metadata`, `auth.cimd`: no remote uses OAuth
- n/a `auth.secret_in_url`: no templated URL

### tool hygiene: 79 (weight 30)

- pass `tools.descriptions`: every tool has a description
- pass `tools.description_length`: descriptions are concise
- pass `tools.schemas`: every tool has an object input schema
- warn `tools.annotations`: only 2 of 4 tools declare readOnlyHint or destructiveHint
- fail `tools.directory_hints`: no tool declares all four of readOnlyHint, destructiveHint, idempotentHint and openWorldHint; missing: promptarch_list_artifact_types (destructiveHint, idempotentHint, openWorldHint); promptarch_enhance_prompt (readOnlyHint, destructiveHint, idempotentHint, openWorldHint); promptarch_lint_artifact (destructiveHint, idempotentHint, openWorldHint); …
- pass `tools.token_cost`: about 1,012 tokens to load every tool
- pass `tools.api_dump`: tools are not a one-to-one API dump

### stability: 100 (weight 15)

- pass `stability.changes`: 2 tool changes in 30 days
- pass `stability.rug_pull`: no tool changed its meaning under the same name

### dependencies: n/a (weight 15)

- n/a `deps.known_vulns`, `deps.mcp_sdk_version`, `deps.resolvable`: no npm or PyPI package

- tools: 4, about 1,012 tokens to load them all
- tool set hash: a88b8bb5baec678c39643e59a85e05c4
- tools last changed: 2026-10-17T03:53:08.780Z
- badge: [![protogrid quality](https://protogrid.dev/badge/ai.promptarch/mcp.svg)](https://protogrid.dev/servers/ai.promptarch/mcp)
- Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed.

### Tool changes

- 2026-10-17 `promptarch_enhance_prompt` added
  - Enhance a raw prompt with PromptArch's Prompt Enhancer: PromptArch asks 1-3 targeted clarifying questions, then rewrites the prompt from the person's answers, optimized for target_model. If this client supports MCP elicitation, the questions appear to the person as a form inside this call. Otherwise call once with just `prompt` to get the questions (free), ask the person, and call again with `questions` and `answers`. Never answer the questions yourself. The enhancement costs 5 credits; questions are free. Requires a PromptArch API key configured as an Authorization: Bearer pk_... header on this MCP server.
- 2026-10-04 `promptarch_lint_artifact` changed (description, inputSchema)
  - before: Lint an AI agent context file (CLAUDE.md, AGENTS.md, Cursor rules, Copilot instructions, memory files) with PromptArch's deterministic linter: ~30 research-backed rule families covering current-model anti-patterns (tool-forcing, reasoning-echo, sycophancy triggers, contradictions), per-tool size budgets, secrets/dangerous commands, and hidden-Unicode injection ("Rules File Backdoor"). Returns grade, 0-100 score, and line-anchored findings. Free: no account or credits needed.
  - after: Lint an AI agent context file (CLAUDE.md, .claude/rules/ files, AGENTS.md, Cursor rules, Copilot instructions, memory files) with PromptArch's deterministic linter: ~30 research-backed rule families covering current-model anti-patterns (tool-forcing, reasoning-echo, sycophancy triggers, contradictions), per-tool size budgets, secrets/dangerous commands, and hidden-Unicode injection ("Rules File Backdoor"). Returns grade, 0-100 score, and line-anchored findings. Free: no account or credits needed.

Full history: https://api.protogrid.dev/v1/servers/ai.promptarch%2Fmcp/changes

## Remotes

- https://mcp.promptarch.ai/mcp (streamable-http, auth none, reachable true, uptime30d 0.95614034)

## Packages

_none_

## Tools (4)

- `promptarch_enhance_prompt`: Enhance a raw prompt with PromptArch's Prompt Enhancer: PromptArch asks 1-3 targeted clarifying questions, then rewrites the prompt from the person's answers, optimized for target_model. If this client supports MCP elicitation, the questions appear to the person as a form inside this call. Otherwise call once with just `prompt` to get the questions (free), ask the person, and call again with `questions` and `answers`. Never answer the questions yourself. The enhancement costs 5 credits; questions are free. Requires a PromptArch API key configured as an Authorization: Bearer pk_... header on this MCP server.
- `promptarch_generate_artifact`: Generate an AI agent context artifact (e.g. context_pack, claude_md) from a project description. Requires a PromptArch API key configured as an Authorization: Bearer pk_... header on this MCP server. Consumes credits.
- `promptarch_lint_artifact`: Lint an AI agent context file (CLAUDE.md, .claude/rules/ files, AGENTS.md, Cursor rules, Copilot instructions, memory files) with PromptArch's deterministic linter: ~30 research-backed rule families covering current-model anti-patterns (tool-forcing, reasoning-echo, sycophancy triggers, contradictions), per-tool size budgets, secrets/dangerous commands, and hidden-Unicode injection ("Rules File Backdoor"). Returns grade, 0-100 score, and line-anchored findings. Free: no account or credits needed.
- `promptarch_list_artifact_types`: List the artifact types promptarch_generate_artifact can produce from a project description.