agentmods.dev

wasmagent-js CLAUDE.md

wasmagent-js CLAUDE.md is an instructions file for coding agents from WasmAgent/wasmagent-js. It costs 4,124 tokens per session, scanned A, original, Apache-2.0.

Development instructions for WasmAgent/wasmagent-js, a project that records evidence from tool-using agents and controls their security policies.

In plain words
What is it for?
Guiding code changes, maintaining repository boundaries, and working with the project's firewall, evidence records, auditing, and agent runtime.
Why use it?
They clarify which parts belong in this repository and which responsibilities belong in other repositories.

Instructions file

not rated 5 6d ago A scan 4,124 tokens original Apache-2.0
View the source file WasmAgent/wasmagent-js
Install

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.

agentmods
npx agentmods add instructions/wasmagent/wasmagent-js/claude-md
Clone the repo
git clone --depth 1 https://github.com/WasmAgent/wasmagent-js

Wrote 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.

agentmods badge for wasmagent-js CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/wasmagent/wasmagent-js/claude-md.svg)](https://agentmods.dev/instructions/wasmagent/wasmagent-js/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/wasmagent/wasmagent-js/claude-md"><img src="https://agentmods.dev/badge/instructions/wasmagent/wasmagent-js/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 4,124 This file is loaded in full into every session.
When invoked 4,124 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.04124 $0.04124
Opus 5 $0.02062 $0.02062
Sonnet 5 $0.00825 $0.00825
Haiku 4.5 $0.00412 $0.00412

Measured 3d ago against content hash a55f2e6b2e19, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

wasmagent-js 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 3d 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.

CLAUDE.md · 305 lines

How it starts

The opening of the file, as written. The whole thing — 305 lines — stays where its author put it; the contents beside it link to each section on GitHub.

wasmagent-js — Development Guide for Claude

What this project is (and is not)

Is: Verifiable evidence layer + security control plane for MCP and tool-using agents. Three public entry points: @wasmagent/mcp-firewall (protect), @wasmagent/aep (record), @wasmagent/compliance (audit+train).

Is NOT — do not implement these:

  • A general-purpose agent framework (LangChain / Mastra / AutoGen territory)
  • A Cursor / Claude Code / Codex competitor (no IDE UX, no remote execution at scale)
  • A compliance certification tool (never claim "satisfies EU AI Act / ISO 42001")
  • A universal RAG / workflow engine
  • A training framework (TRL / Axolotl territory) — we produce training data, not training code

Repository Boundaries

This runtime is one repo in a tree with sharp edges. Do not re-implement what another repo owns; call it instead.

This repository owns

  • The agent runtime — kernels (QuickJS/Pyodide/Wasmtime/Remote), executor, orchestration (agent.run(), supervisor, subagents, handoff, teams)
  • MCP firewall / gateway / attestation — process-level policy enforcement
  • AEP evidence emission (@wasmagent/aep) — recording signed evidence at runtime
  • Model adapters, smartrouter, tools, and the developer-facing SDK surface

Other repositories own — do not duplicate here

Capability Owner
AEP + compliance schema definitions (ConstraintIR, task-spec, rollout-wire, etc.) wasmagent-protocol (@wasmagent/protocol)
Symbolic verification engine (CEL / wazero sandbox / Z3 SMT) symkernel (HTTP service)
Gateway-level HTTP evidence (Proxy-Wasm) wasmagent-proxy
Training-data pipeline trace-pipeline
AgentBOM / MCP Posture / Trust Passport agent-trust-infra / open-agent-audit

Allowed cross-repo patterns

  • Schemas: consume @wasmagent/protocol — never copy or inline schema JSON into this repo. Need a change? Open it against wasmagent-protocol.
  • Verification: heavy symbolic checks (CEL/Z3/hard-isolation sandbox) call symkernel over HTTP via the Criterion/ConstraintIR protocol. Keep a light inline fast path here; the authority is symkernel.
  • This runtime emits AEP records; it does not define their schema and does not run the verification engine.

Read the full file on GitHub · 305 lines

Changes

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.

  1. 3d ago First seen · 305 lines · 4,124 tokens per session scan A a55f2e6b2e19

Subscribe to this mod's changes

wasmagent-js CLAUDE.md is an instructions file published in the GitHub repository WasmAgent/wasmagent-js (5 stars, last pushed 6d ago), licensed Apache-2.0. It adds 4,124 tokens to every session, about $0.0206 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.

LLMagentcloudflare-workerscompliancemcpsandbox
Related

Other instructions, from other repositories

agentos CLAUDE.md

Claude Code instructions for rivet-dev/agentos, covering agentos, boundaries, security model, sqlite schema ownership and runtime and registry.

rivet-dev/agentos · 4,057 tokens

agent-sandbox AGENTS.md

Instructions for kubernetes-sigs/agent-sandbox, covering agents.md, project summary, repository layout, agent skills and build, test, lint.

kubernetes-sigs/agent-sandbox · 4,218 tokens

agent-sandbox copilot-instructions.md

Instructions for kubernetes-sigs/agent-sandbox: Project Context & Architecture: Refer to AGENTS.md for full project background, module layout, toolchain versions, and core conventions.

kubernetes-sigs/agent-sandbox · 293 tokens

ironcurtain CLAUDE.md

Instructions for provos/ironcurtain, covering claude.md, general workflow, git & worktrees, git workflow and platform considerations.

provos/ironcurtain · 5,773 tokens

ironcurtain copilot-instructions.md

Instructions for provos/ironcurtain, covering ironcurtain copilot review instructions, platform support, security invariants, ai sdk v6 api (breaking changes) and code conventions.

provos/ironcurtain · 850 tokens

ironcurtain security-boundary.instructions.md

Instructions for provos/ironcurtain, covering security boundary review rules, mandatory checks and structural invariant order.

provos/ironcurtain · 747 tokens