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/fizzbee-io/fizzbee/claude-mdgit clone --depth 1 https://github.com/fizzbee-io/fizzbeeWrote 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/fizzbee-io/fizzbee/claude-md)<a href="https://agentmods.dev/instructions/fizzbee-io/fizzbee/claude-md"><img src="https://agentmods.dev/badge/instructions/fizzbee-io/fizzbee/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 | $0.00455 | $0.00455 |
| Opus 5 | $0.00228 | $0.00228 |
| Sonnet 5 | $0.00091 | $0.00091 |
| Haiku 4.5 | $0.00046 | $0.00046 |
Grade A, and why
fizzbee CLAUDE.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.
What it actually says
Claude Code Instructions for Fizzbee
Build Commands
- Build the model checker:
bazel build //:fizzbee - Run the model checker:
./fizz path/to/spec/file.fizz(use the wrapper script, not the bazel binary directly)
Important Guidelines
- Do NOT use gazelle to update BUILD files
- The
./fizzwrapper script handles compilation and execution - always use it instead of./bazel-bin/fizzbee_/fizzbee
Testing
- Reference examples are in
examples/references/organized by feature - Many examples have
baseline.txtfiles for comparing output metrics (Nodes, Valid Nodes, Unique States, PASSED/FAILED status) - Generated files (
.jsonAST files andout/directories) can be deleted after testing
Running the test suite
python3 test_references.py
Runs all 94 reference examples (those with a baseline.txt) and all lovable specs in examples/requirements/lovable/. Compares output semantically (strips timestamps, elapsed times, file paths). Reports PASS/FAIL per example and a summary.
To update a baseline after an intentional change:
./fizz examples/references/NN-name/Spec.fizz > examples/references/NN-name/baseline.txt 2>&1
Project Structure
lib/- Core library code (starlark types, symmetry, JSON marshalling)modelchecker/- Model checking engine (processor, threads, cloning, state visitor)parser/- Fizz language parserproto/- Protocol buffer definitionsexamples/- Example specificationsreferences/- Reference examples organized by feature (01-xx through 16-xx)tutorials/- To be removed. Old tutorial specifications, some of which may be outdated and may not even work
samples/- Larger sample specifications. Again, not organized and may be outdated and may not even work, but kept for reference. These are not merged into git.
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 · 42 lines · 455 tokens per session scan A 7bfd9d4b2dd6
fizzbee CLAUDE.md is an instructions file published in the GitHub repository fizzbee-io/fizzbee (347 stars, last pushed 8d ago), licensed Apache-2.0. It adds 455 tokens to every session, about $0.0023 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
vera AGENTS.md
Instructions for aallan/vera, covering agents.md — instructions for ai agents, for agents writing vera code, conformance programs as reference, workflow and commands.
emilia-protocol CLAUDE.md
Instructions for emiliaprotocol/emilia-protocol, covering emilia protocol - repo guide, required context, critical, build & ship and outbound & claims.
emilia-protocol copilot-instructions.md
Instructions for emiliaprotocol/emilia-protocol: Read AGENTS.md and AICONTEXT.md before proposing changes. Current evidence and provenance are in public/.well-known/emilia-context.json. Treat archived, staged, private, and outreach documents as non-authoritative for current behavior or standards status.
emilia-protocol GEMINI.md
Instructions for emiliaprotocol/emilia-protocol: Read AGENTS.md and AICONTEXT.md before analyzing or editing this repository. Use public/.well-known/emilia-context.json for current evidence counts and provenance. Do not infer current behavior or standards status from archived, staged, private, or randomly sampled…
jacobian AGENTS.md
Instructions for morluto/jacobian, covering jacobian agent guide, what we are building, non-negotiable boundaries, implement mathematics directly and mathematical boundedness is a proof obligation.
aristo CLAUDE.md
Instructions for aretta-ai/aristo, covering claude.md — aristo working agreement, §1. commit size — small or medium only, §2. commit messages — semantic / conventional, §3. changelog.md — one line per commit, in the same commit and §4. test-first — no test, no claim of correctness.