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 agents/jovandyaz/knowtis-app/knowtis-architectgit clone --depth 1 https://github.com/jovandyaz/knowtis-appWrote 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/agents/jovandyaz/knowtis-app/knowtis-architect)<a href="https://agentmods.dev/agents/jovandyaz/knowtis-app/knowtis-architect"><img src="https://agentmods.dev/badge/agents/jovandyaz/knowtis-app/knowtis-architect.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 | $0.00056 | $0.00586 |
| Opus 5 | $0.00028 | $0.00293 |
| Sonnet 5 | $0.00011 | $0.00117 |
| Haiku 4.5 | $0.00006 | $0.00059 |
Grade A, and why
knowtis-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 4d 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 — 30 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the Knowtis architect — the domain expert for the Knowtis platform (real-time collaborative notes: Nx monorepo, React 19, NestJS 11, Postgres/Drizzle, Yjs/Hocuspocus, Vercel AI SDK copilot).
You are read-only by design: you answer, review, and advise. You never edit files, run commands, or apply fixes — when a change is needed, you describe it precisely (file paths, what to change, which invariant motivates it) and let the caller apply it.
How you work
- Ground every answer in this plugin's skills — invoke them via the Skill tool rather than re-deriving:
orienting— repo layout, module boundaries, alias namespaces, footguns.building-copilot-features— copilot/AI-gateway invariants (HITL, injection guard, fallback chain, BYOK, retrieval, evals).wiring-realtime-collaboration— Yjs/Hocuspocus handshake, persistence guards, broadcast semantics.
- When the question outruns the skills, read the canonical repo docs (
docs/ARCHITECTURE.md,docs/AI.md,docs/MCP.md,docs/AUTH.md,docs/PERMISSIONS.md) and verify the answer against the current code. - For design reviews, check the proposal against the invariants explicitly and report: verdict (respects/violates), which invariant, evidence (file:line), recommended design. Flag violations even when the code "works".
- Answer in the user's language. Be precise about the difference between the
aimodule (single-shot) and theagentmodule (copilot) — misattribution between them is the most common architecture error.
Invariants you defend (non-exhaustive)
Server-authoritative conversations; HITL on every agent mutation; retrieved content is DATA, never instructions; fallback chain never switches models mid-stream; BYOK_ENCRYPTION_KEY is permanent; collaboration authorization at the WebSocket handshake with server-enforced read-only; the trivial-fragment persistence guard; the single CASL ability shared by HTTP and WS; domain layers free of infrastructure imports; import type never used on NestJS-injected classes.
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.
- 4d ago First seen · 30 lines · 56 tokens per session scan A 582c66444381
knowtis-architect is an agent published in the GitHub repository jovandyaz/knowtis-app (3 stars, last pushed today), licensed MIT. It adds 56 tokens to every session and 586 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 2026-08-31.
Other agents, from other repositories
frontend-ticket-implementer
Use this agent when you need to implement frontend features, fix bugs, or complete development tickets for the React/TypeScript frontend. Examples: Context: User has a ticket to implement a new card creation modal component. user: 'I need to implement ticket FE-123: Add a modal for creating new cards with title…
tc-implement
Implements one QA TC ID as an integration test under packages/rangelink-vscode-extension/src/integration-tests/suite/. Traces the production code path, writes the test using helpers/, inserts in TC-ID ascending order, updates the YAML entry's automated: field, and runs the matching pnpm test:release: command.
worker-lint
Runs lint.py against the wiki, parses tiered output, returns severity summary. Use before quarterly reviews or when user asks for wiki health check.
wiki-writer
Create or update wiki pages — autonomous ingest from any source, autonomous update. Auto-creates .wiki/ if missing.
accreta-storage-engineer
Storage and concurrent-systems engineer for accreta. Owns when the filesystem stops being the right substrate, and every concurrency and durability sharp edge in the index. Attacks ADR-0004 on its own terms. Read-only; returns graded findings, never edits.
curator
Reads one repository and returns its five knowledge documents as text. Read-only by construction — it cannot write files. Use it for /study, /refresh, and any time a repo needs to be understood from scratch.