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 Koredotcom/agents-mcp-tools --skill arch-platformgit clone --depth 1 https://github.com/Koredotcom/agents-mcp-toolsWrote 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/koredotcom/agents-mcp-tools/arch-platform)<a href="https://agentmods.dev/skills/koredotcom/agents-mcp-tools/arch-platform"><img src="https://agentmods.dev/badge/skills/koredotcom/agents-mcp-tools/arch-platform/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/koredotcom/agents-mcp-tools/arch-platform"><img src="https://agentmods.dev/badge/skills/koredotcom/agents-mcp-tools/arch-platform.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00027 | $0.01336 |
| Opus 5.5 | $0.00011 | $0.00534 |
| Sonnet 5.5 | $0.00005 | $0.00267 |
| Haiku 4.5 | $0.00003 | $0.00134 |
Grade A, and why
arch-platform 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 1mo 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 — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Arch Platform Operations
Use Arch as the Agent Platform control plane. Preserve the user's requested scope and obtain authorization before material external mutations.
Source of Truth
- Read
arch://guidance/v1/manifestand the relevant feature or tool detail. - Require
schemaVersionto equal"1"before using catalog dependencies, confidence, or planning prompts. If the manifest is absent or incompatible, stop catalog-driven planning and use only legacy tool discovery and existing project-builder operations; do not infer dependencies. - Treat schema-derived operations as executable only when support is
implementedorverified. - Treat
authoritative-livereadiness as project-specific; inspect it through existing project-builder resources before planning project mutations. - Documentation-only or unsupported entries do not authorize an operation.
Operating Rules
- Resolve every
requiresdependency before mutation. - Use opaque auth-profile, integration, and MCP-server references; never place raw secrets in ordinary tool arguments.
- Respect
read,write,idempotent_write,grant_gated_write, anddestructive_writesafety. Treat plainwriteas non-idempotent unless the live operation proves otherwise. - Before grant-gated or destructive work, show the exact target and obtain the required confirmation/grant.
- If a mutation has an unknown outcome, verify it; never retry blindly.
- After acting, supply the operation's
verificationRequiredContext, call itsvalidatesWithtool/action, and compare the result withverificationExpectedEvidence. If the guidance says the validation read cannot prove the original outcome, report that limitation and rely only on the non-replayed original result. - Do not claim support for platform features absent from the public Arch operation catalog.
Whole-project sequence for any MCP client
Read debug_docs topic mcp/project-build-playbook for the complete client-neutral path;
the connected Studio topic is guides/arch-mcp-project-build. In order: connect and confirm
workspace; create/read the project; configure model and secure AuthProfiles; create/test
integrations and MCP servers; author tables, tools and workflows; create reusable prompts,
feedback templates and external-agent entries; create/read/guardedly edit agents and pin
prompt versions; configure optional policy, alerts and channels; publish immutable agent
versions; create/read the typed deployment; run evals and a real channel session. Use the
catalog's validatesWith after every mutation. platform_versions:create and promote are
unsupported legacy actions; use publish and deployment promotion. platform_arch_sop is an
alternative guided creation path, and platform_arch_auto_loop is for later governed repair.
Provider-specific channel provisioning is outside the current direct MCP surface.
What ships with it
1 file 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.
- 1mo ago Changed · +64 lines 357009285da7
- 2mo ago First seen · 29 lines · 27 tokens per session scan A 4d58243697e7
arch-platform is a skill published in the GitHub repository Koredotcom/agents-mcp-tools (4 stars, last pushed 1mo ago), licensed MIT. It adds 27 tokens to every session and 1,336 once invoked, about $0.0001 per session on Opus 5.5. 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.
arch-platform compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| arch-platformKoredotcom/agents-mcp-tools | Skill | 4 | 1mo ago | MIT | A | 27 | Codex |
| imagegenopenai/codex | Skill | 135k | today | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 100k | today | Apache-2.0 | A | 19 | Claude Code |
| browser-automationopenclaw/openclaw | Skill | 394k | today | MIT | A | 36 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 80k | 6d ago | MIT | A | 200 | Claude Code, Codex |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 95 | Claude Code |
| chat-pet-sprite-creationmicrosoft/vscode | Skill | 195k | today | MIT | A | 53 | Claude Code, Codex |