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 skills add cdeust/ai-architect-mcp-spec --skill skillgit clone --depth 1 https://github.com/cdeust/ai-architect-mcp-specWrote 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/skills/cdeust/ai-architect-mcp-spec/skill)<a href="https://agentmods.dev/skills/cdeust/ai-architect-mcp-spec/skill"><img src="https://agentmods.dev/badge/skills/cdeust/ai-architect-mcp-spec/skill.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 5.1 | $0.00109 | $0.05053 |
| Opus 5 | $0.00055 | $0.02527 |
| Sonnet 5 | $0.00022 | $0.01011 |
| Haiku 4.5 | $0.00011 | $0.00505 |
Grade B, and why
ai-architect-mcp-spec scanned grade B with 1 finding 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 8d 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.
Asks the agent to reveal its instructionsmediumSystem prompt leakage
Directions to print, repeat or translate the system prompt extract configuration the operator did not intend to expose.
**Execute:** Issue **all** invocations in a **single message** with **multiple parallel Agent tool calls**. (Sequential dispatch produces a correct result but multiplies wall-clock time by N — for self-check batches with How it starts
The opening of the file, as written. The whole thing — 383 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Architect MCP Spec (v0.8.0) — Dispatcher Protocol
You (the host) drive a loop:
- Call
start_pipeline→ receive an envelope with{ run_id, messages, action, ... }. - Display every entry in
messagesto the user. - Execute
actionper the dispatch table below. - Call
submit_action_result(run_id, result)→ receive next envelope. - Repeat until
action.kind === "done"oraction.kind === "failed".
emit_message is never returned to the host as action. The runner coalesces all status messages into the messages array; the action field always carries something the host actually has to do.
ENVELOPE SHAPE
Every response from start_pipeline and submit_action_result has this shape:
{
"run_id": "run_abc_123",
"current_step": "context_detection",
"messages": [
{ "text": "🟢 PRD Spec Generator — TRIAL TIER\n...", "level": "info" }
],
"action": { "kind": "ask_user" | "call_pipeline_tool" | ... , ... },
"state_summary": { "sections": [...], "clarification_rounds": 3, "errors": 0 }
}
run_id— handle forsubmit_action_resultandget_pipeline_state. Caputre once from the first response and reuse.messages— banners/status lines collected while the runner advanced internally to reachaction. May be empty. Display eachtextat the givenlevel(defaultinfo) before executingaction.action— what you must execute. Neveremit_message. Always one of:ask_user,call_pipeline_tool,call_cortex_tool,spawn_subagents,write_file,done,failed.
TERMINOLOGY (read this before the dispatch table)
| Term | Means |
|---|---|
| the host | Claude Code (or another MCP-aware client) running this dispatcher loop |
| the project directory | The codebase root where .mcp.json lives — distinct from "the host" |
| the runner | The MCP server's stateless reducer behind the pipeline tools |
| the result (always typed) | The ActionResult value the host passes to submit_action_result — exactly one of four kinds |
| agent output | The raw text returned by an Agent tool call to a subagent — appears in subagent_batch_result.responses[i].raw_text |
a judge (purpose: "judge") |
A spawned subagent invocation that returns a JudgeVerdict JSON object. The same word also names the judge field inside JudgeVerdict, which holds the AgentIdentity of the agent that rendered the verdict. |
| routing token | One of correlation_id, invocation_id, batch_id, question_id — opaque strings the host MUST echo back unchanged |
messages |
The array of { text, level } entries the runner collected while advancing internally to the substantive action. Display each text at its level before executing action. |
| substantive action | Any action kind that requires host execution and a submitted result: ask_user, call_pipeline_tool, call_cortex_tool, spawn_subagents, write_file. Plus terminal kinds done and failed. (emit_message is NEVER returned to the host — the runner coalesces it.) |
| coalescing | The runner's behavior of collecting status signals (which would otherwise be individual emit_message actions) into the messages array, so each response carries exactly one substantive action plus its accumulated message context. |
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 8d ago First seen · 383 lines · 109 tokens per session scan B e569e5d74e49
ai-architect-mcp-spec is a skill published in the GitHub repository cdeust/ai-architect-mcp-spec (3 stars, last pushed yesterday), licensed MIT. It adds 109 tokens to every session and 5,053 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it B with 1 finding (asks the agent to reveal its instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
adeu-redlining
Use this skill when reviewing, editing, redlining, or negotiating an existing Microsoft Word document (.docx) — including proposing edits as tracked changes, accepting or rejecting existing tracked changes, replying to comments, comparing two versions, sanitizing author metadata, or finalizing a contract for…
lain
Structural code intelligence for AI coding agents. Use this skill when the user wants to understand how a codebase is organized (modules, call graphs, file dependencies), find where to start reading, trace the impact of a change, find code by meaning, or understand what a symbol does in its full context. Do NOT use…
wiki
Rebuild a flow-first project wiki using the mimirs wiki MCP tool. Use when the user asks to generate, rebuild, refresh, or write the wiki for a codebase.
plan
Design an implementation plan before writing code — where the change lands, what it will touch, what could break, and the steps in order. Use when asked to plan a feature, scope a change, or figure out how to approach an edit before making it. To assess a change that already exists (a diff, refactor, or rename), use…
research
Answer a hard, open-ended question about how the project works or is built by synthesizing every source — code, structure, git history, prior decisions, discussion, caveats — and verifying each claim against the source. Use for deep cross-cutting questions that span more than one area. Narrower siblings — a single…
scout
Research the web for solutions, competitors, alternatives, or prior art for a decision facing this project — then store the findings in project memory so they survive the session. Use when choosing a library or approach, comparing tools, checking what others do, or evaluating whether to build vs adopt.