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 instructions/eva-ics/tabularium/agents-mdgit clone --depth 1 https://github.com/eva-ics/tabulariumWhat 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 | $0.01098 | $0.01098 |
| Opus 5 | $0.00549 | $0.00549 |
| Sonnet 5 | $0.00220 | $0.00220 |
| Haiku 4.5 | $0.00110 | $0.00110 |
Grade A, and why
tabularium AGENTS.md scanned grade A with 1 finding 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 2d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- Example config: [config.toml.example](config.toml.example) — copy to `config.toml` for a writable local path, or rely on `just run` (uses the example when `config.toml` is absent). API scrolls: [docs/json-rpc-methods.m How it starts
The opening of the file, as written. The whole thing — 40 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent notes (tabularium)
For machine spirits and maintainers: layout, commands, and doctrine.
Coding standards: the Tabularium MCP scroll kb/coding-rules must be applied for new code blocks when architecture or conventions are uncertain.
Layout
- Workspace root: Cargo.toml — members
tabularium/,tabularium-server/,tabularium-cli/. - Library crate: tabularium/ — Cargo features
db(SQLite + Tantivy +Database),client(rpc::Client+reqwest; impliesdb),full= both. Default =client(same as full stack). Server depends ondbonly; CLI onclient. Seejust check-tabularium-db/just check-tabularium-core. - Server binary: tabularium-server/ (
tabularium-server), Axum REST/api/doc/*+/api/search+ JSON-RPCPOST /rpc. - CLI binary: tabularium-cli/ (
tb). - Example config: config.toml.example — copy to
config.tomlfor a writable local path, or rely onjust run(uses the example whenconfig.tomlis absent). API scrolls: docs/json-rpc-methods.md, docs/rest-api.md, docs/curl-examples.md. - Optional Python checks: requirements-dev.txt, tests/ (
pytest tests/) — setTABULARIUM_TEST_URLwhen a server is running; for CLI tests, setTABULARIUM_TB_BINor installtbonPATH. - Selenium / Web UI (
@pytest.mark.webui): do not run againstlocalhostor127.0.0.1. Tests are skipped unlessTABULARIUM_TEST_URLorTABULARIUM_URLpoints at host10.90.1.122(same port as the running server). Use the integration host for all Web UI browser tests;just ci-testbinds the ephemeral server on0.0.0.0:<port>and sets the base URL tohttp://10.90.1.122:<port>automatically. - Docs: docs/. Meetings/tasks: tabularium MCP
Behaviour notes (meeting consensus)
- Index / DB drift: if the process dies between a SQLite write and a Tantivy commit, the index can disagree with the DB. Stage 1 accepts this; a future
just/admin rebuild hook is the intended remedy. accessed_at/touch: every successful logical read viaDatabase::get_documentcallsStorage::touchafter content is resolved (cache hit or miss). Iftouchfails, the error is propagated to the caller (noOkwith stale silence).- Read cache:
mokaasync cache holds document bodies only;cache_size == 0keeps the cache object but never stores entries. Mutations invalidate the affected document id after a successful storage write. reindex:Database::reindex(None)clears the Tantivy index and rebuilds from everydocumentsrow.Database::reindex(Some(...))re-upserts only rows in the targeted directory subtree via the internal directory row identifier; other indexed documents are left as-is. Public API uses directory/path vocabulary only.- Tracing: the library uses
tracingspans onSqliteDatabase::init,Databasefaçade methods,SqliteStorage(Storage), andSearchIndexI/O. Install a subscriber in binaries/tests (e.g.tracing_subscriber) and setRUST_LOG=tabularium=debug(ortrace) to see them; the library crate does not initialize global logging.
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.
- 2d ago First seen · 40 lines · 1,098 tokens per session scan A e594b14a8fd9
tabularium AGENTS.md is an instructions file published in the GitHub repository eva-ics/tabularium (22 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 1,098 tokens to every session, about $0.0055 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.