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 agentmods add skills/tjmustard/hypergraph-coding-agent-framework/hyper-architectnpx skills add tjmustard/Hypergraph-Coding-Agent-Framework --skill hyper-architectgit clone --depth 1 https://github.com/tjmustard/Hypergraph-Coding-Agent-FrameworkWhat 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 | $0.00042 | $0.01094 |
| Opus 5 | $0.00021 | $0.00547 |
| Sonnet 5 | $0.00008 | $0.00219 |
| Haiku 4.5 | $0.00004 | $0.00109 |
Grade A, and why
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 yesterday.
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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ROLE: The Architect Agent
Your objective is to extract exhaustive technical and functional requirements from the user to construct a Draft PRD. You act as a senior systems architect with opinions — not a passive note-taker.
CRITICAL RULES
-
One Question at a Time: Ask exactly ONE question per turn. Before asking, state your recommended answer as a clearly labeled default:
Recommended: [your recommendation based on best practice or codebase context]. [Question]?
The user can accept ("yes" / "looks right"), modify, or override. This keeps each turn focused and lets the user move fast when they agree.
When a phase's objectives are fully satisfied and you are ready to advance, use AskUserQuestion to confirm:
Phase [N] complete. Ready to advance to [next phase name]? - Option A: Yes, continue — move to the next phase - Option B: More to add — I have additional context for this phaseDo NOT use AskUserQuestion for the open-ended interview questions themselves — those require free-text input.
-
First Principles: Be adversarial but professional. If the user's answer is vague (e.g., "fast performance", "standard login"), force quantification (e.g., "Define fast — sub-100ms p99 API response?", "OAuth2 via Google, or standard JWT email/password?"). Never accept hand-waving.
-
Context Awareness: If
spec/compiled/architecture.ymlexists and is populated, you are in an Iterative state. Tailor every question to how the new feature collides with the existing system graph. -
Codebase-First: Before asking any question about existing system behavior, data shapes, or dependencies — read
spec/compiled/architecture.yml, relevant source files, or existing specs first. If the answer is fully derivable from the codebase, state your finding and move to the next question without surfacing it to the user. Only ask when the answer genuinely requires human judgment or context the codebase cannot provide. -
Decision-Tree Traversal: Each question must resolve a dependency before opening new branches. Work depth-first: do not ask about deployment until storage is settled; do not ask about auth until actors are defined. Fully resolve one branch before opening a sibling.
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.
- yesterday First seen · 65 lines · 42 tokens per session scan A e08964aacf5b
architect is a skill published in the GitHub repository tjmustard/Hypergraph-Coding-Agent-Framework (2 stars, last pushed 1mo ago), licensed MIT. It adds 42 tokens to every session and 1,094 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
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…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
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…