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/sazardev/shiki/claude-mdgit clone --depth 1 https://github.com/sazardev/shikiWrote 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/sazardev/shiki/claude-md)<a href="https://agentmods.dev/instructions/sazardev/shiki/claude-md"><img src="https://agentmods.dev/badge/instructions/sazardev/shiki/claude-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 | $0.37682 | $0.37682 |
| Opus 5 | $0.18841 | $0.18841 |
| Sonnet 5 | $0.07536 | $0.07536 |
| Haiku 4.5 | $0.03768 | $0.03768 |
Grade A, and why
shiki CLAUDE.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 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
`whisper-blas-bin-*.zip`), with the model (`ggml-*.bin`) downloaded once via `curl`/`wget` from How it starts
The opening of the file, as written. The whole thing — 1,808 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Project
shiki (私記) is a TUI note-taking app written in Rust, inspired by Yazi's three-pane
layout and modal navigation. Notes are plain Markdown files with YAML frontmatter,
organized into "notebooks" (directories, each its own git repo). Full design spec,
motivation, keybindings, config format, and included themes are documented in IDEA.md —
read it before making architectural changes, since it's the source of truth for intended
behavior (layout, CLI commands, config schema, etc).
Commands
cargo build --workspace # build everything
cargo check --workspace # fast type-check (use this while iterating)
cargo clippy --workspace --all-targets # lint; keep this clean before considering work done
cargo fmt --all # format (run after editing, before checking clippy)
cargo run -p shiki-cli -- <args> # run the binary, e.g. `-- new "titulo"`, `-- daily`, no args launches the TUI
There are ~293 #[test]s: 134 in shiki-core, 17 in shiki-config, 129 in shiki-tui, 13 in
shiki-cli — cargo test --workspace is green. They're all inline #[cfg(test)] modules inside
the source files (no tests/ dirs, no #[ignore], no fixture setup), so the pattern set by
panel_drawer::tests (shiki-tui/src/panel_drawer.rs) — covering drawer_hit_at's mouse
coordinate math as a plain function of numbers, not &App — is the norm. When adding tests,
put them as #[cfg(test)] modules in the relevant file; shiki-core/shiki-config have no
TUI/terminal dependency, so they're the easiest to unit test. For shiki-tui, prefer designing
the function to not need &App in the first place (as drawer_hit_at does) over constructing a
full App in a test.
To exercise the CLI without touching the real user config/data, override XDG dirs (used via
directories::ProjectDirs::from("", "", "shiki") in shiki-config):
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 · 1,808 lines · 37,682 tokens per session scan A 3c1827d2f109
shiki CLAUDE.md is an instructions file published in the GitHub repository sazardev/shiki (291 stars, last pushed yesterday), licensed MIT. It adds 37,682 tokens to every session, about $0.1884 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
kimun CLAUDE.md
Instructions for nico2sh/kimun, covering kimun, docs and rules.
reflect-open AGENTS.md
AGENTS.md instructions for team-reflect/reflect-open, covering purpose, what is reflect, product principles, agent workflow and development workflow.
siyuan AGENTS.md
AGENTS.md instructions for siyuan-note/siyuan, covering agents.md, 1. non-negotiable constraints, do not hand-edit, verification and prohibited operations and 2. project-specific rules.
reflect-open CLAUDE.md
Claude Code instructions for team-reflect/reflect-open: See AGENTS.md for the full project overview, tech stack, database tables, code conventions, and development cycle.
zen CLAUDE.md
Claude Code instructions for sheshbabu/zen, covering claude.md, development commands, build commands, dependencies and go build tags.
hippocampus CLAUDE.md
Instructions for sturlese/hippocampus, covering layout, frontmatter schema, conventions, answering questions from the vault (read order) and contradictions.