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/buildonspark/spark/claude-mdgit clone --depth 1 https://github.com/buildonspark/sparkWrote 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/buildonspark/spark/claude-md)<a href="https://agentmods.dev/instructions/buildonspark/spark/claude-md"><img src="https://agentmods.dev/badge/instructions/buildonspark/spark/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 | $0.01821 | $0.01821 |
| Opus 5 | $0.00911 | $0.00911 |
| Sonnet 5 | $0.00364 | $0.00364 |
| Haiku 4.5 | $0.00182 | $0.00182 |
Grade A, and why
spark 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 yesterday.
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 — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Spark Repository
Spark is a Bitcoin Layer 2 protocol enabling instant, low-cost Bitcoin transactions using FROST threshold signatures.
For deep protocol and architecture context, see .claude/CLAUDE.md.
Setup
mise trust && mise install
lefthook install # git hooks (optional)
Components
spark/- Go backend (Signing Operators) - seespark/CLAUDE.mdsdks/js/- JavaScript/TypeScript SDK - seesdks/js/packages/spark-sdk/CLAUDE.mdsigner/- Rust FROST signing implementationlrc20.dev/- BTKN token layer (Rust)protos/- Protocol buffer definitions
Code Quality Guidelines
Comments
Be judicious with comments. Prefer clear naming and well-structured functions over explanatory comments.
Avoid redundant comments:
// Bad - comment just repeats the code
// Verify operator signatures
verifyOperatorSignatures(sigs)
// Good - no comment needed, function name is clear
verifyOperatorSignatures(sigs)
When comments ARE useful:
- Non-obvious "why" explanations (e.g., race conditions, protocol requirements)
- Public API documentation
- Security-critical sections
- References to external specs or papers
Comment quality guidelines:
- Explain "why", not "what" - the code shows what happens; comments explain reasoning
- Comments must be self-contained - never reference "the old implementation" since future readers won't have that context
- Only add non-obvious information -
// Save to databasebeforedb.Save()wastes space - Don't comment out code - delete it; git remembers
- A Jira ticket belongs in a comment as
TODO(SP-XXXX), or appended as a citation to a comment that already reads on its own - never as a bare "See SP-1234" standing in for the reason
Logging (Go)
We use Zap (go.uber.org/zap) for all logging, replacing log/slog from the standard library. Zap has two logger types:
SugaredLogger— default for logs with dynamic values. Use printf-style formatting vialogger.Sugar().Logger— use only when attaching approved, searchable attributes.
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.
- yesterday First seen · 163 lines · 1,821 tokens per session scan A 5d9bf52f45f1
spark CLAUDE.md is an instructions file published in the GitHub repository buildonspark/spark (110 stars, last pushed 4d ago), licensed Apache-2.0. It adds 1,821 tokens to every session, about $0.0091 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 2026-09-03.
Other instructions, from other repositories
ton-rs AGENTS.md
AGENTS.md instructions for ston-fi/ton-rs, covering ton-rs agent guide, repository, boundaries, public api and ton invariants.
bsv-wallet-cli CLAUDE.md
Claude Code instructions for Calhooon/bsv-wallet-cli, covering bsv-wallet-cli, architecture, conventions, development and build.
cardano-init CLAUDE.md
Claude Code instructions for input-output-hk/cardano-init, covering claude.md, project overview, commands, build and run.
StellarX-Workshop-Template-2026 CLAUDE.md
Instructions for ALGOREX-PH/StellarX-Workshop-Template-2026, covering stellarx workshop starter — project notes for ai tools, stack / versions, testnet reference, stellar gotchas (these waste the most time) and where things live.
allora-chain AGENTS.md
AGENTS.md instructions for allora-network/allora-chain, covering agents.md, quick rules (apply to most changes), go conventions, changes that require extra steps and minimal context strategy.
allora-chain CLAUDE.md
Claude Code instructions for allora-network/allora-chain, covering claude.md, quick rules (apply to most changes), go conventions, changes that require extra steps and minimal context strategy.