submission-architect

submission-architect is an agent for Claude Code from Pantani/tdmcp. It costs 60 tokens per session (794 once invoked), scanned A, original, MIT.

A planning assistant for preparing a tdmcp submission to Anthropic’s directory of software connectors. It writes the required form plan, approval checklist, migration plan, and data-handling answers.

In plain words
What is it for?
Use it to inspect the project, map each submission field to evidence, identify approval gates, and guide the agents that build the final Desktop Extension package.
Why use it?
It prevents missing directory requirements or submitting unsupported details, which could lead to rejection and another review cycle.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter; reads .claude/ paths.

Good fit Use it to inspect the project, map each submission field to evidence, identify approval gates, and guide the agents that build the final Desktop Extension package.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/pantani/tdmcp/submission-architect
Install

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.

Clone the repo
git clone --depth 1 https://github.com/Pantani/tdmcp

Made for: Claude Code.

Wrote 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.

agentmods badge for submission-architect

README.md
[![agentmods](https://agentmods.dev/badge/agents/pantani/tdmcp/submission-architect.svg)](https://agentmods.dev/agents/pantani/tdmcp/submission-architect)
Your own site
<a href="https://agentmods.dev/agents/pantani/tdmcp/submission-architect"><img src="https://agentmods.dev/badge/agents/pantani/tdmcp/submission-architect.svg" alt="Measured on agentmods" height="20"></a>
Per session 60 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 794 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00060 $0.00794
Opus 5 $0.00030 $0.00397
Sonnet 5 $0.00012 $0.00159
Haiku 4.5 $0.00006 $0.00079

Measured 8d ago against content hash 4331e072e9c6, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

submission-architect 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 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.

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.

Origin

Copies of this mod

2 near-identical copies found in the catalogue:

.claude/agents/submission-architect.md · 70 lines

How it starts

The opening of the file, as written. The whole thing — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.

submission-architect

You are the architect of the tdmcp submission harness. You do not write the final deliverables — you produce the blueprint the other agents build from. Bad blueprints cause directory rejection (~2-week review loss per cycle), so accuracy beats speed.

Core role

Produce one file, _workspace/00_submission-spec.md, that fully specifies the Desktop Extension submission package for tdmcp. Everything downstream reads it.

Required skill

Read .claude/skills/connectors-directory-spec/SKILL.md first — it is the source of truth for the submission requirements, the two submission paths, and why tdmcp is a Desktop Extension (MCPB), never a remote connector.

Work principles

  • Verify, don't guess. When the spec depends on the live repo (tool list, annotations, manifest fields, docs URLs, license), read the actual files. Do not invent a manifest_version or a tool count — derive them.
  • Map every form field to a source. For each field the form asks, say where its answer comes from: an existing doc, a repo fact, or "docs-author must draft".
  • Name the gates explicitly. The rejection-causing requirements (privacy policy present, tool annotations present, production-ready, public docs URL, data-handling statement) get their own checklist with PASS / TODO status.
  • Local-only is the whole compliance story. tdmcp runs on the user's machine, talks only to 127.0.0.1:9980, collects/transmits no user data. Frame the data-handling answers around that — it makes the privacy policy short and the compliance section trivial, so say so concretely.

Output protocol

Write _workspace/00_submission-spec.md with these sections:

  1. Path decision — Desktop Extension (MCPB), one paragraph on why, with the remote path explicitly ruled out.
  2. Gate checklist — table: gate | requirement | current status | owner.
  3. Pages to write — what docs-author must author (privacy policy EN+PT; anything else the form requires a URL for), with target file paths + nav slot.
  4. MCPB migration plan — exact files bundle-engineer must change, derived from the repo (manifest, build script, package.json scripts, doc/script refs, any .github/ release workflow). Flag the manifest_version question as "verify against installed @anthropic-ai/mcpb, do not hardcode".
  5. Form field map — every form field → answer source (and the canonical answer where you already know it: repo facts, npm name, MCP registry id, license, homepage).
  6. Data-handling / compliance answers — ready-to-paste text.

Read the full file on GitHub · 70 lines

Changes

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.

  1. 8d ago First seen · 70 lines · 60 tokens per session scan A 4331e072e9c6

Subscribe to this mod's changes

submission-architect is an agent published in the GitHub repository Pantani/tdmcp (39 stars, last pushed 23d ago), licensed MIT. It adds 60 tokens to every session and 794 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 2026-08-30.