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 skills/orphograph/orphograph/openclawnpx skills add Orphograph/Orphograph --skill openclawgit clone --depth 1 https://github.com/Orphograph/OrphographWhat 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 | $0.00096 | $0.00505 |
| Opus 5 | $0.00048 | $0.00253 |
| Sonnet 5 | $0.00019 | $0.00101 |
| Haiku 4.5 | $0.00010 | $0.00051 |
Grade A, and why
orphograph-anchor 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.
What it actually says
Orphograph Anchor — agent audit trail
You have orpho_agent_anchor.py in this skill's directory. It sends
only hashes to orphograph.com — file contents never leave the machine.
A receipt proves the exact bytes existed at anchor time; it does not prove
authorship, ownership, or that the action was correct. Never claim more.
Auth comes from ORPHO_API_KEY or ORPHO_PACK_TOKEN in the environment.
If neither is set, tell the operator to get a key at
https://orphograph.com and stop — do not retry.
When and how
1. After a consequential action — pipe a one-line record:
echo "2026-07-10T14:02Z sent invoice #42 to <recipient>" | \
python3 orpho_agent_anchor.py anchor-text --label "action:invoice-42"
2. Daily memory snapshot (cron/heartbeat) — anchors one canonical
manifest of every *.md memory file, so silent history edits are
detectable later:
python3 orpho_agent_anchor.py anchor-memory ~/workspace --label "memory-daily"
3. Skill/prompt provenance — before sharing a skill file, anchor it and include the receipt id when you share; after receiving one, verify:
python3 orpho_agent_anchor.py anchor-file ./SKILL.md --label "skill:my-skill v3"
python3 orpho_agent_anchor.py verify <receipt_id> --file ./received_skill.md
Rules
- Receipts append to
.orphograph/receipts.jsonl— never delete or edit it. - On
http_error402/429, stop and notify the operator (out of credits or rate-limited); do not loop. --dry-runfirst if unsure whether a file should be anchored at all.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 49 lines · 96 tokens per session scan A 33f2806e0b9e
orphograph-anchor is a skill published in the GitHub repository Orphograph/Orphograph (0 stars, last pushed yesterday), licensed MIT. It adds 96 tokens to every session and 505 once invoked, about $0.0005 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 2026-08-31.
Other skills, from other repositories
binance-tokenized-securities-info
Query Ondo tokenized US stock data on Binance Web3. Covers: supported stock token list, RWA metadata (company info, attestation reports), market and per-asset trading status (with corporate action codes for earnings, dividends, splits), real-time on-chain data (token price, holders, circulating supply, market cap), US…
payment-assistant
Binance Pay Assistant - Send and Receive crypto payments. Send: QR code payment from Funding Wallet (C2C + PIX). Use when user wants to buy/purchase/pay/transfer/send, confirm/cancel payment, or query order status. Requires QR code data. PIX QR codes (pix, br.gov.bcb.pix) are auto-detected. Receive: Generate QR codes…
ccxt-csharp
CCXT cryptocurrency exchange library for C# and .NET developers. Covers both REST API (standard) and WebSocket API (real-time). Helps install CCXT, connect to exchanges, fetch market data, place orders, stream live tickers/orderbooks, handle authentication, and manage errors in .NET projects. Use when working with…
new-exchange
Scaffold a new CCXT exchange integration in TypeScript, following the certified-exchange template. Walks through describe(), required unified methods, parsers, capability flags, sandbox setup, and static fixtures. Use when adding support for an exchange that does not exist yet under ts/src/.
fiat
Query Binance fiat payment capabilities — supported countries, currencies, payment methods, limits, and crypto prices — via public APIs, plus authenticated order/payment history lookup. Use whenever users ask about buying or selling crypto with fiat, depositing or withdrawing fiat, fiat-crypto exchange rates, payment…
meme-rush
Live launchpad feed + AI hot topics for meme tokens. (1) meme-rush: real-time lifecycle feed on launchpads (Pump.fun, Four.meme) — brand-new launches, currently-finalizing / bonding-curve tokens, and just-migrated-to-DEX tokens; filter by dev behavior, age, market cap. (2) topic-rush: AI-detected hot market narratives…