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/cammac-creator/ibanforge/claude-mdgit clone --depth 1 https://github.com/cammac-creator/ibanforgeWrote 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/cammac-creator/ibanforge/claude-md)<a href="https://agentmods.dev/instructions/cammac-creator/ibanforge/claude-md"><img src="https://agentmods.dev/badge/instructions/cammac-creator/ibanforge/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 5.1 | $0.02835 | $0.02835 |
| Opus 5 | $0.01418 | $0.01418 |
| Sonnet 5 | $0.00567 | $0.00567 |
| Haiku 4.5 | $0.00283 | $0.00283 |
Grade B, and why
ibanforge CLAUDE.md scanned grade B 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 5d 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
`find .claude/worktrees/<nom> -type f -not -path "*/node_modules/*" -newermt "-1 day"`. How it starts
The opening of the file, as written. The whole thing — 213 lines — stays where its author put it; the contents beside it link to each section on GitHub.
IBANforge
API de validation IBAN et lookup BIC/SWIFT avec micropaiements x402, interface MCP pour agents AI, données SEPA/VoP, classification émetteur (vIBAN detection), et indicateurs de risque compliance.
Stack
- Runtime : Node.js 22+ / TypeScript (20 est en fin de vie depuis le 2026-04-30 ; 24 est impossible tant que better-sqlite3 11 ne publie pas de binaire linux-x64 pour l'ABI 137)
- Framework : Hono
- Database : SQLite (better-sqlite3) —
data/bic.sqlite(121k+ BIC entries from GLEIF + SwiftCodes/MIT + SIX + EBA Step2 SCT + Bundesbank + NBP, plus 1,100+ Swiss clearing entries SIX — counts drift at each monthly refresh, read them live viagetEntryCount()/getChClearingCount()),data/stats.sqlite - Payments : x402/hono (USDC micropayments)
- AI Agents : MCP SDK (Model Context Protocol)
- Deploy : Docker → Railway
- Domain : ibanforge.com
Architecture
src/
index.ts # Entry point — Hono app + server
types.ts # Shared TypeScript types
routes/
iban-validate.ts # POST /v1/iban/validate (single IBAN)
iban-batch.ts # POST /v1/iban/batch (up to 100)
bic-lookup.ts # GET /v1/bic/:code (BIC/SWIFT lookup)
ch-clearing.ts # GET /v1/ch/clearing/:iid (Swiss BC-Nummer lookup)
health.ts # GET /health
stats.ts # GET /stats
landing.ts # GET / (HTML landing page)
demo.ts # GET /v1/demo (free examples)
lib/
iban.ts # IBAN validation logic (mod97, BBAN parsing)
enrich.ts # Post-validation enrichment (BIC, issuer, SEPA, risk, CH clearing)
ch-clearing.ts # Swiss BC-Nummer lookup, institution type detection
issuers.ts # EMI/neobank classification (85 known BIC8 mappings — the live count is served in /llms.txt)
bic-validator.ts # BIC format validation (ISO 9362)
bic-lookup.ts # BIC database queries
countries.ts # ISO country data (IBAN lengths, BBAN structures, SEPA zones, VoP, country risk)
stats.ts # Stats recording and queries
db.ts # Database connections
middleware/
x402.ts # x402 payment middleware
mcp/
server.ts # MCP server (validate_iban, batch_validate, lookup_bic, lookup_ch_clearing)
db/
schema.sql # SQLite schema
seed.ts # GLEIF BIC-LEI data seeder
scripts/
enrich-countries.ts # Backfill country_name on existing data
seed-bc-nummer.ts # Download + seed SIX BankMaster CSV into bic.sqlite
data/
bic.sqlite # Pre-built BIC + Swiss clearing database (tracked in git)
stats.sqlite # API usage stats
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.
- 5d ago First seen · 213 lines · 2,835 tokens per session scan B 2403aba21822
ibanforge CLAUDE.md is an instructions file published in the GitHub repository cammac-creator/ibanforge (3 stars, last pushed today), licensed MIT. It adds 2,835 tokens to every session, about $0.0142 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
openbanking-mcp CLAUDE.md
Claude Code instructions for Beennnn/openbanking-mcp, covering le fournisseur de données : enable banking, la règle qui gouverne tout le reste, /.local/share/bankread/ledger/ n'est pas un cache, ce qui ne se fait jamais ici and où vit quoi.
bharatpay-mcp CLAUDE.md
Claude Code instructions for angelina10504/bharatpay-mcp, covering claude.md, what this is, layout, engineering standards and 1. every tool has a pure-function offline validator.
qontoctl CLAUDE.md
Claude Code instructions for alexey-pelykh/qontoctl, covering qontoctl — claude instructions, project overview, repository structure, package dependency graph and development commands.
fli CLAUDE.md
Instructions for punitarani/fli, covering claude.md, project overview, development commands, core development tasks and install dependencies.
fli AGENTS.md
Instructions for punitarani/fli, covering agents.md, cursor cloud specific instructions, development commands, testing caveats and releasing.
anytype-mcp CLAUDE.md
Instructions for anyproto/anytype-mcp, covering claude.md, common commands, development, testing and code quality.