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/alinalihassan/lesma/agents-mdgit clone --depth 1 https://github.com/alinalihassan/LesmaWrote 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/alinalihassan/lesma/agents-md)<a href="https://agentmods.dev/instructions/alinalihassan/lesma/agents-md"><img src="https://agentmods.dev/badge/instructions/alinalihassan/lesma/agents-md.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 | $0.03672 | $0.03672 |
| Opus 5 | $0.01836 | $0.01836 |
| Sonnet 5 | $0.00734 | $0.00734 |
| Haiku 4.5 | $0.00367 | $0.00367 |
Grade A, and why
Lesma 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 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 — 174 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — Lesma project guide for AI agents
This file gives future agents (and humans) a quick reference for how the Lesma compiler works, how to build it, and how to validate changes.
What Lesma is
Lesma is a compiled, statically typed, imperative, object-oriented language that targets LLVM. The compiler is C++23 and uses LLVM for code generation, optimization, JIT, and object-file emission. The standard library is written in Lesma (.les files in src/stdlib/).
Project layout
src/— Compiler and stdlibsrc/cli/main.cpp— CLI entry point (whenLESMA_BUILD_CLIis on).src/liblesma/— Core library:Common/— Utils, logging, errors (LesmaError,CodegenErrorinBackend/CodegenError.h).Frontend/— Lexer, Parser (tokens → AST).AST/— AST node definitions and visitor interface.Token/— Token types andTokenclass.Symbol/— Symbol table,Type,Value,TypeUtils.Backend/— Codegen (AST → LLVM IR),MangleUtils,CodegenTypeUtils, linking/JIT.
src/stdlib/— Lesma standard library (e.g.base.les,math.les,time.les).src/lsp/—lesma-lsp(whenLESMA_BUILD_LSPis ON). Advertises UTF-8 position encoding (PositionEncodingKind::UTF8); clients must sendPosition.characteras UTF-8 code units (bytes) from the line start. Enables vcpkg featurelsp(libgit2, minimal:pcre2only, no HTTPS/SSH) for workspace-wide.lesdiscovery in Git work trees without shelling out togit.
tests/lesma/success/— Programs that must compile and run (exit 0).tests/lesma/failure/— Programs that must be rejected (expected to fail).scripts/run_tests.sh— Runs the compiler on all success/failure cases (run + compile for each).benchmark/— Google Benchmark targets plus thebenchmark suiteintegration harness (benchmark/SuiteWallClock.cpp).
Wall-clock benchmark (benchmark suite)
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 · 174 lines · 3,672 tokens per session scan A bca47fcf9011
Lesma AGENTS.md is an instructions file published in the GitHub repository alinalihassan/Lesma (21 stars, last pushed 4mo ago), licensed MIT. It adds 3,672 tokens to every session, about $0.0184 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 instructions, from other repositories
kara CLAUDE.md
Claude Code instructions for karalang/kara, covering claude.md, commands, branch management, claiming a bug (multi-session coordination) and architecture.
Rynix AGENTS.md
AGENTS.md instructions for Ali-Rashidi-80/Rynix, covering guide for ai agents working on rynix, canonical toolchain, honesty rules and layout.
sigil-lang CLAUDE.md
Claude Code instructions for Daemoniorum-LLC/sigil-lang, covering sigil language - agent guide, the canonical compiler, run sigil programs, run tests and why rust compiler?.
morloc CLAUDE.md
Instructions for morloc-project/morloc, covering claude.md, project overview, general rules, checking code and other rules.
cuda-metal AGENTS.md
AGENTS.md instructions for Lulzx/cuda-metal, covering agents.md — cumetal development guide, project identity, current focus, repository structure and non-negotiable constraints.
hexa-lang CLAUDE.md
Instructions for dancinlab/hexa-lang, covering hexa-lang, project, tree, performance gold standard and release integrity — absolute top guardrail.