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/trustless-ai/agent-sdk/add-ercnpx skills add trustless-ai/agent-sdk --skill add-ercgit clone --depth 1 https://github.com/trustless-ai/agent-sdkWrote 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/skills/trustless-ai/agent-sdk/add-erc)<a href="https://agentmods.dev/skills/trustless-ai/agent-sdk/add-erc"><img src="https://agentmods.dev/badge/skills/trustless-ai/agent-sdk/add-erc.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 | $0.00066 | $0.10043 |
| Opus 5 | $0.00033 | $0.05021 |
| Sonnet 5 | $0.00013 | $0.02009 |
| Haiku 4.5 | $0.00007 | $0.01004 |
Grade A, and why
add-erc scanned grade A 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 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
* Read-only contract calls: methods return `Result<T, ClientError>` where fetching happens via `self.provider.fetch(key)`. No `send`/broadcast in core — write methods belong in the `providers` crate or a separate host-on How it starts
The opening of the file, as written. The whole thing — 400 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Add ERC
Generate off-chain client support for one ERC that agent-sdk doesn't yet implement.
The output has two layers:
- Layer 1 — contract wrappers:
client.ts/client.pythat call the on-chain contract via an RPC node (read/send/verify). - Layer 2 — pure recompute:
recompute.ts/recompute.pywith stateless functions that reproduce the ERC's deterministic computations from public inputs, verified against golden conformance vectors without requiring any blockchain node.
Process
-
Determine which ERC. If not specified, ask which ERC (by number, e.g. "ERC-8301") to add. If the
agent-ercssubmodule should be read from something other than its currently checked-outmain, ask for the branch, tag, commit, or local path to use, and check that out inagent-ercs/before continuing (default: whatevermaincurrently points to). -
Read the spec. Read the ERC's interface file(s) and
README.mdunderagent-ercs/contracts/<category>/<ERCXXXX>/. -
Classify recompute-to-verify capability. An ERC can make more than one distinct claim — classify each central claim separately rather than forcing one verdict for the whole ERC. For each claim, determine: can a caller independently obtain the same authoritative answer without trusting whoever originally submitted it — either by recomputing off-chain from public data, or by calling a deterministic, callable-by-anyone, immutable on-chain check themselves — or does the guarantee terminate at trusting a specific deployment's unfixed convention (a signing scheme, a derived value) with no way to even know what to check? Write out the reasoning — this is the part of the job that isn't mechanical.
- See
typescript/src/identity/ERC8004/README.mdfor a clean NOT-verifiable case: the interface leaves a signing convention completely unfixed, so there's nothing a generic SDK can check at all. - See
typescript/src/verify/ERC8274/README.mdfor a split verdict on one ERC: the core validity check is recompute-to-verify (anyone can call the deployed, immutable verifier contract themselves and get an authoritative answer — that's a real instance of recompute-to-verify, not "just asking the same contract again"), while a separate derived value (an audit-trail digest) is NOT, because one of its inputs isn't exposed anywhere in the interface. Don't let one claim's verdict force the other's.
- See
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 · 400 lines · 66 tokens per session scan A a264e800691f
add-erc is a skill published in the GitHub repository trustless-ai/agent-sdk (2 stars, last pushed 2d ago), licensed Apache-2.0. It adds 66 tokens to every session and 10,043 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
adding-personhog-rpc
Guide for adding a new RPC to personhog-replica and personhog-router. Covers eligibility checks, proto definition, code generation for Python and Node.js clients, Rust implementation (storage trait, postgres queries, service handler, router wiring), and index compatibility validation. Use when adding a new gRPC…
langgraph-docs
Fetches and references LangGraph Python documentation to build stateful agents, create multi-agent workflows, and implement human-in-the-loop patterns. Use when the user asks about LangGraph, graph agents, state machines, agent orchestration, LangGraph API, or needs LangGraph implementation guidance.
stripe-projects
Use after E2B sandbox/API access has been provisioned through Stripe Projects and the user needs to use the resulting E2B API key with the E2B CLI, JavaScript SDK, Python SDK, or Code Interpreter SDK.
dd-code-generation
Use pup CLI for immediate Datadog operations or generate code for integration into applications.
gjc-sdk-author
Author trusted-local TypeScript and Python scripts that operate GJC sessions through the broker-bound CLI.
ark-sdk-development
Regenerate and debug types across the ARK stack (SDK, API, Dashboard). Use when fixing TypeScript type errors in ark-dashboard, updating types after CRD changes, regenerating types.ts from OpenAPI spec, debugging "Property does not exist on type" schema errors, or adding custom SDK functionality via overlays. Covers…