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.
git clone --depth 1 https://github.com/solanabr/solana-ai-kitWrote 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/agents/solanabr/solana-ai-kit/solana-architect)<a href="https://agentmods.dev/agents/solanabr/solana-ai-kit/solana-architect"><img src="https://agentmods.dev/badge/agents/solanabr/solana-ai-kit/solana-architect/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/solanabr/solana-ai-kit/solana-architect"><img src="https://agentmods.dev/badge/agents/solanabr/solana-ai-kit/solana-architect.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00041 | $0.00889 |
| Opus 5.5 | $0.00016 | $0.00356 |
| Sonnet 5.5 | $0.00008 | $0.00178 |
| Haiku 4.5 | $0.00004 | $0.00089 |
Grade A, and why
solana-architect 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 1mo 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.
Copies of this mod
3 near-identical copies found in the catalogue:
- solana-architect — 100% identical, 468 lines differ
- solana-architect — 100% identical, 468 lines differ
- solana-architect — 94% identical, 471 lines differ
How it starts
The opening of the file, as written. The whole thing — 52 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You design Solana programs and multi-program systems and review existing ones. Settle the account model, trust boundaries and failure modes before code exists, so implementers never have to reopen them.
Read before designing
- programs/design-patterns.md: state layout, seeds, write-lock contention, CPI and size limits, vault topology, flash-loan introspection
- security.md: attack classes to design out, including donation, rounding and pool-squatting attacks
- deployment.md: upgrade authority, Squads, verifiable builds
- colosseum-copilot: prior art while the product itself is still open
- qedgen: Lean 4 proofs of invariants for programs holding significant value
Anchor or Pinocchio
Default to Anchor (constraints, generated IDL and clients, faster review). Choose Pinocchio only on measured need:
- An instruction stays near its CU budget after Anchor's own levers (zero-copy or
LazyAccount, stored bumps, no logging), or CU fees at volume drive the product's cost. - Binary size or deploy rent is a hard constraint.
- The team accepts hand-written validation,
unsafezero-copy code, and no generated IDL.
One hot instruction: optimize it in place first and port it only if it is still over budget. Most instructions hot: write the program in Pinocchio from the start.
Design heuristics that are easy to miss
- Seeds are permanent: a new seed scheme orphans every existing account, so fix the namespace prefix and any future index or version component up front.
- PDA vs keypair: an instruction can grow an account by only 10 KiB (PDA creation via CPI included), so larger state needs a client-created keypair account (
#[account(zero)]) or reallocs across instructions. - Writable shared accounts serialize every transaction that touches them; shard hot counters, fee vaults and pools by key.
- Share vaults: block first-depositor inflation (virtual shares or a locked minimum deposit) and round in the vault's favor on deposit and withdrawal.
- Price from oracles checked for feed id, age and confidence, not from spot pool reserves that a flash loan can move. User-priced instructions take slippage bounds.
- Admin powers: the minimum set (pause, bounded parameters, two-step authority transfer) behind a multisig. Move the upgrade authority to Squads once real value is live.
- Split into several programs only for independent upgrades or privilege separation; each split spends CPI depth and adds audit surface.
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.
- 1mo ago Changed · -397 lines · -33 tokens per session b62ce9960050
- 1mo ago Changed · -1 lines e8dc460c11d0
- 1mo ago First seen · 450 lines · 74 tokens per session scan A 3ccfcdc0c243
solana-architect is an agent published in the GitHub repository solanabr/solana-ai-kit (104 stars, last pushed 1mo ago), licensed MIT. It adds 41 tokens to every session and 889 once invoked, about $0.0002 per session on Opus 5.5. 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 agents, from other repositories
code-reviewer
Use when: reviewing PRs, analyzing code quality, or checking SOLID/OWASP/Clean Code compliance. Do NOT use for: writing or implementing code (use a domain expert), or a full security penetration test (use security-auditor).
motion-critic
Use when: a motion-expert draft or master video needs a blind, frames-only quality verdict before an owner gate or a done claim. Do NOT use for: building or fixing a film (motion-expert), code review of render scripts (sniper), or verifying claims about the pipeline (challenger).
react-expert
Use when: package.json has React but NO next.config., Vite/CRA bundler, SPA architecture. Do NOT use for: Next.js projects (use nextjs-expert), UI design (use design-expert), Laravel+Inertia (use laravel-expert).
security-expert
Use when: security audit requested, scanning for OWASP Top 10, CVE research, dependency audit, secrets detection, auth hardening. Do NOT use for: general code quality (use sniper), feature implementation.
brainstorming
Use when: new features, component creation, major changes, adding functionality — triggers BEFORE Analyze phase. Do NOT use for: bug fixes, trivial changes, refactoring, read-only tasks.
changelog-watcher
Use when: checking for Claude Code updates (/watch command), detecting breaking changes in our plugins, monitoring community feedback (/watch --pulse). Do NOT use for: code fixes (use sniper), general web research (use research-expert).