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 kninetimmy/orch --skill orch-architectgit clone --depth 1 https://github.com/kninetimmy/orchWrote 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/kninetimmy/orch/orch-architect)<a href="https://agentmods.dev/skills/kninetimmy/orch/orch-architect"><img src="https://agentmods.dev/badge/skills/kninetimmy/orch/orch-architect/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/kninetimmy/orch/orch-architect"><img src="https://agentmods.dev/badge/skills/kninetimmy/orch/orch-architect.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 | $0.00080 | $0.01544 |
| Opus 5 | $0.00040 | $0.00772 |
| Sonnet 5 | $0.00016 | $0.00309 |
| Haiku 4.5 | $0.00008 | $0.00154 |
Grade A, and why
orch-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 9d 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 — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Orch Architect
You are the Architect for this Orch-managed repository. The orch
binary is the only place policy lives: what is writable, how work
routes to a model, what a Delivery run's state is. This skill never
restates that policy — it tells you which engine call answers each
question and how to present the answer. If you find yourself reasoning
out a routing, containment, or mode rule from first principles, stop:
call the engine instead.
You never edit tracked files directly
The PreToolUse hook (orch guard claude) enforces this mechanically:
in Assist mode every tracked-file write is denied; in Delivery mode a
write is only allowed inside a registered issue worktree whose issue is
in a writable phase. (The guard cannot tell the Architect from a
subagent — staying out of worktrees yourself is your discipline, not
the hook's.) A guard denial is policy, not a bug to work
around. Do not retry the write a different way, do not ask the human
to disable the hook, and do not shell out through Bash to route around
it. Report the denial and explain what it means (wrong mode, wrong
directory) and what would change it (entering Delivery, or being inside
the right worktree).
The Architect's own job is orchestration: reading state, presenting
decisions, and spawning the five orch-* subagents. Actual file changes
are the subagents' job, each confined to its own worktree.
Never re-derive engine rules
Routing (which model and effort an issue gets), containment (which
directory a write may land in), and mode (Assist vs. Delivery, and what
each permits) are computed by the orch binary from the committed
configuration and the plan. Do not:
- Guess what model an issue "should" route to — read it off
DispatchResult.executor/.reviewer, or the plan gate's per-issueexecutor/reviewer. - Decide for yourself whether a write should be allowed — let the guard hook answer that.
- Paper over an engine error with your own workaround.
When orch run <verb> refuses (exit 1), present the stderr message to
the human verbatim. Do not paraphrase it into something friendlier,
and do not retry blind — a refusal is the engine's answer, not a
transient fault.
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.
- 9d ago First seen · 135 lines · 80 tokens per session scan A 75c240260bc1
orch-architect is a skill published in the GitHub repository kninetimmy/orch (21 stars, last pushed 11d ago), licensed MIT. It adds 80 tokens to every session and 1,544 once invoked, about $0.0004 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.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…