Borrowing it
Nothing to install: this file belongs to MasRama/nara. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/MasRama/nara/main/AGENTS.mdgit clone --depth 1 https://github.com/MasRama/naraWrote 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/instructions/masrama/nara/agents-md)<a href="https://agentmods.dev/instructions/masrama/nara/agents-md"><img src="https://agentmods.dev/badge/instructions/masrama/nara/agents-md/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/instructions/masrama/nara/agents-md"><img src="https://agentmods.dev/badge/instructions/masrama/nara/agents-md.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 | $0.01533 | $0.01533 |
| Opus 5.5 | $0.00613 | $0.00613 |
| Sonnet 5.5 | $0.00307 | $0.00307 |
| Haiku 4.5 | $0.00153 | $0.00153 |
Grade A, and why
nara AGENTS.md 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 27d 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 — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Nara is an architecture-aware TypeScript application kit. Build by feature, not by layer.
Stack: TypeScript, Node.js 22+, Hono + @hono/node-server, Vue 3 + Vite + vue-router, SQLite (better-sqlite3, raw SQL), Zod, Vitest. Session auth owned by the auth feature.
Authority: user instruction → this file → ARCHITECTURE.md → tests → implementation. Git history explains past decisions; it never overrides current code.
Architecture model
src/features/<feature>/ contract.ts · index.ts · server/ · web/ (optional) · tests/
src/app/ server.ts · router.ts · App.vue · pages/ · layouts/
src/shared/ config/ · database/ · logging/ · realtime/ · security/ · storage/
resources/app.ts thin Vite entry mounting the app shell
src/features/<feature>/index.tsis the general/server-facing public boundary. Cross-feature server use imports only from there.src/features/<feature>/web/index.tsis the optional browser-safe boundary.src/app/imports browser surfaces only from there.- Internals (
server/*,web/pages/*,web/components/*,web/client) are private. Feature dependencies must be acyclic. - Details:
docs/feature-model.md,ARCHITECTURE.md.
Hard rules
- Feature-first: new capability →
src/features/<name>/(skeleton vianara make feature). Never add globalcontrollers//services//repositories/trees. - Boundaries: never import another feature's
server/*,web/pages/*,web/components/*, orweb/client— use itsindex.ts/web/index.ts. Never export server-only symbols throughweb/index.ts. - Browser code under
web/must not importserver/files,@/shared/database, Node-only built-ins, or server-only packages. - Server is authoritative: enforce auth/permissions in Hono routes, never only in Vue. Every
/apiroute declares its access with a guard (signedIn/allow) orpublicRoute; the server refuses to start otherwise. Permission slugs are<resource>.<action>;adminbypasses where the route requires it. - Responses use
{ success: true, message, data? }/{ success: false, message, code, errors? }, English messages, Zod validation at the route boundary viajsonInput/queryInput(401 auth, 403 permission, 404 absent, 409 conflict, 422 validation). Edits of a shared record send therevisionthey were based on; an older one gets 409STALE_REVISIONcarryingcurrent, which the form merges instead of overwriting. - Interface text is English and lives in the page's template. The UI shows API refusals by
messageand branches oncode, never on message text; every code a route sends is declared in its Feature contract (<FEATURE>_REFUSAL_CODES, plusAPI_REFUSAL_CODESfromsrc/shared/security/codes.ts). - Tests outside Auth create accounts and grants through
src/app/tests/personas.ts, never by writing Auth's tables. - SQL lives in the owning feature's repository via
better-sqlite3prepared statements; multi-write replacements use transactions. No ORM, no string-interpolated values. - Locked stack: do not replace Hono, add a frontend framework (React/Svelte/Nuxt/SSR), add a native HTTP engine (Ultimate Express/uWebSockets.js), or wrap Hono/Vue behind a custom Nara abstraction. New dependency genuinely required → prefer the existing stack or standard library and update the actual package manifest (
package.json). If it changes current architecture, updateARCHITECTURE.md; otherwise keep rationale close to the code or test that enforces it. - No overengineering: no speculative abstractions, plugin systems, caches, DI containers, RPC/ORM/validation frameworks, or duplicated architecture metadata. Keep changes scoped; no mass-formatting, no unrelated refactors, no secrets, no force-push.
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.
- 27d ago Changed · +28 tokens per session e3eafa8fb8a2
- 28d ago Changed · +2 lines · +172 tokens per session 3bb6988a4e1f
- 29d ago Changed · +1 lines · +65 tokens per session 9ac36a601471
- 1mo ago Changed · -15 lines · -225 tokens per session 0bd210892434
- 2mo ago Changed · -2 lines · -13 tokens per session 6173e20a7be6
- 2mo ago Changed · +2 tokens per session a53076c9b41f
- 2mo ago Changed · -181 lines · -2,432 tokens per session 38341c3aa688
- 2mo ago First seen · 281 lines · 3,936 tokens per session scan A e797ddd43346
nara AGENTS.md is an instructions file published in the GitHub repository MasRama/nara (5 stars, last pushed 27d ago), licensed MIT. It adds 1,533 tokens to every session, about $0.0061 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.
nara AGENTS.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| nara AGENTS.mdMasRama/nara | Instructions file | 5 | 27d ago | MIT | A | 1,533 | Codex, OpenCode |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | yesterday | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | yesterday | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 274k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |