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/claudioemmanuel/squeez/claude-mdgit clone --depth 1 https://github.com/claudioemmanuel/squeezWrote 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/claudioemmanuel/squeez/claude-md)<a href="https://agentmods.dev/instructions/claudioemmanuel/squeez/claude-md"><img src="https://agentmods.dev/badge/instructions/claudioemmanuel/squeez/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 | $0.01456 | $0.01456 |
| Opus 5 | $0.00728 | $0.00728 |
| Sonnet 5 | $0.00291 | $0.00291 |
| Haiku 4.5 | $0.00146 | $0.00146 |
Grade A, and why
squeez 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 3d 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.
curl (local build), npm (pre-built download), cargo install. Release workflow builds universal macOS (lipo), Linux x86_64/aarch64 musl, Windows MSVC on tag push. How it starts
The opening of the file, as written. The whole thing — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Guidance for Claude Code working in this repo.
Commands
cargo test # all unit tests; `cargo test <name>` for one
cargo build --release
bash build.sh # build + install to ~/.claude/squeez/bin/ + register hooks
# NOTE: re-registers every host adapter, not just Claude Code
bash bench/run.sh # filter-mode benchmarks
python3 bench/verify_tokens.py # real-tokenizer check (tiktoken cl100k_base) vs chars/4
./target/release/squeez benchmark [--json|--baseline|--efficiency-proof]
No Makefile — tooling is Cargo-native.
Hard constraints
- Zero runtime dependencies.
Cargo.tomllists onlylibc(Unix signal forwarding). Binary size and CI reproducibility depend on stdlib-only builds. squeez setupmust never shell out to an interpreter. It usedpython3 -cuntil that broke setup on machines withoutpython3on PATH, notably Windows (#190). Usesrc/hosts/settings_json.rs(pure Rust onjson_util).- Every commit needs DCO sign-off — CI enforces via
check-signoff:git commit -s -m "feat(scope): description".
Architecture
Hook-based bash-output compressor for seven CLI agent hosts (Claude Code, Copilot CLI, OpenCode, Gemini CLI, Codex CLI, Pi, Hermes). Intercepts tool invocations and compresses output before the model sees it.
Claude Code hooks: PreToolUse → wrap / budget / prompt-compress / agent-spawn ceiling; SessionStart → init; PostToolUse → track-result + updatedToolOutput rewrite; SubagentStop → feed sub-agent output into SessionContext + release a spawn slot; PreCompact / PostCompact → log + re-arm.
Pipeline (per invocation): smart_filter (ANSI/progress/spinners/timestamps) → dedup (repeats → [×N]) → grouping (≥5 siblings → dir/ N modified) → truncation (head/tail by handler).
Handler dispatch: src/filter.rs detects command type → one of 13+ handlers in src/commands/. extract_name() strips wrappers (npx, bunx, pnpm exec, yarn exec) first. To add one: implement in src/commands/, register in src/commands/mod.rs, add a dispatch arm in src/filter.rs.
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.
- 3d ago First seen · 75 lines · 1,456 tokens per session scan A 107c520924e6
squeez CLAUDE.md is an instructions file published in the GitHub repository claudioemmanuel/squeez (190 stars, last pushed 12d ago), licensed Apache-2.0. It adds 1,456 tokens to every session, about $0.0073 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
plannotator AGENTS.md
AGENTS.md instructions for backnotprop/plannotator, covering plannotator, project structure, server runtimes, installation and environment variables.
solo AGENTS.md
Instructions for solo-agent/solo, covering project testing rules and project service lifecycle rules.
Accordion AGENTS.md
Instructions for a-Fig/Accordion, a project described as: 🏆 AI Hackathon 2026 @ UC Berkeley Intelligent context management for developers.
untether GEMINI.md
Instructions for littlebearapps/untether, covering untether — gemini instructions, stack & conventions, architecture, key rules and testing.
agent-tools AGENTS.md
Instructions for kairyou/agent-tools, covering agent instructions, build and test, layout, settled decisions and docs.
silo AGENTS.md
Instructions for silo-code/silo, covering silo — project guide for coding agents, engineering principles, self-documentation — keep docs in sync as you build, planning a substantial change and domain language — keep the glossary in sync as you build.