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/caspel26/shimwire/claude-mdgit clone --depth 1 https://github.com/caspel26/shimwireWrote 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/caspel26/shimwire/claude-md)<a href="https://agentmods.dev/instructions/caspel26/shimwire/claude-md"><img src="https://agentmods.dev/badge/instructions/caspel26/shimwire/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 | $0.01075 | $0.01075 |
| Opus 5 | $0.00537 | $0.00537 |
| Sonnet 5 | $0.00215 | $0.00215 |
| Haiku 4.5 | $0.00108 | $0.00108 |
Grade A, and why
shimwire 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 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.
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.
How it starts
The opening of the file, as written. The whole thing — 33 lines — stays where its author put it; the contents beside it link to each section on GitHub.
shimwire
A CLI tool with two modes sharing one engine:
- Mock mode: generates a fake-but-schema-valid API server from an OpenAPI spec
- Client mode: a scriptable, git-friendly HTTP test runner (collections as version-controlled TOML files, not a proprietary cloud format)
Full design and phased rollout: shimwire-implementation-plan.md. Read it before making architectural decisions — it defines the stack, project structure, data formats, and phase exit criteria.
Status
Phases 0–3 done. shimwire init, run (Phase 1), mock (Phase 2), and generate (Phase 3 — auto-scaffolds a collection from an OpenAPI spec, with heuristic depends_on linking between a resource's POST and its /{id} operations, auth pre-fill from securitySchemes (bearer/basic/apiKey), and fake request bodies via the Phase 2 schema faker; unresolvable guesses are written as review-comment headers in the generated TOML) are all implemented and tested. Phase 4 (TUI) is explicitly deferred — don't build it unprompted. Phase 5 (polish/distribution) is partial: released as v0.1.0 and published on npm (npm install -g shimwire / bunx shimwire, no build step — ships raw TS off the #!/usr/bin/env bun shebang in src/cli.ts, so bin must stay executable, i.e. chmod +x, and files in package.json must keep scoping the tarball to src/); standalone binaries and Homebrew aren't started. Check what's actually in a directory before assuming the plan's full structure is populated.
Repo is public. main requires the test CI check to pass before merging a PR (direct pushes from the maintainer still bypass this by design — see branch protection settings, not tracked in this repo).
Stack
- TypeScript on Bun (native TS execution,
bun test,bun build --compilefor single-binary distribution) commanderfor CLI parsing,fastifyfor the mock server,@apidevtools/swagger-parserfor OpenAPI,@faker-js/fakerfor fake data,smol-tomlfor collection/env files,picocolorsfor output- No Node-only tooling assumptions — everything should run via
bun, notnode/npm, unless a dependency truly requires it. The one deliberate exception:.github/workflows/publish.yml's actualnpm publishstep uses the real npm CLI viaactions/setup-node, notbun publish—bun publishdidn't pick up theNODE_AUTH_TOKEN-style auth in CI (failed with "missing authentication" despite a correctly-set secret), so don't "fix" that back tobun publishwithout solving the auth problem first.
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 · 33 lines · 1,075 tokens per session scan A daa6cec986c9
shimwire CLAUDE.md is an instructions file published in the GitHub repository caspel26/shimwire (2 stars, last pushed 22d ago), licensed MIT. It adds 1,075 tokens to every session, about $0.0054 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 instructions, from other repositories
api-testing-mcp CLAUDE.md
Instructions for cocaxcode/api-testing-mcp, covering claude.md — @cocaxcode/api-testing-mcp, project overview, token-optimized responses (v0.13+), stack and architecture.
mockserver-monorepo AGENTS.md
AGENTS.md instructions for mock-server/mockserver-monorepo, covering mockserver — agent instructions, instruction priority, project overview, local development environment and project documentation.
gini-agent AGENTS.md
AGENTS.md instructions for Open-Curiosity/gini-agent, covering gini agent instructions, shape, adrs, boundaries and branches.
bread CLAUDE.md
Instructions for tjakoen/bread, covering claude.md — start here, what this is, start here (reading order), this repo is the stack's control plane — run it and non-negotiables (see conventions for the full rules).
agentconfig.org AGENTS.md
Instructions for agentconfig/agentconfig.org, covering agent instructions for agentconfig.org, project overview, target audience, site structure and tech stack.
rosetta command-versions-vs-presence.instructions.md
Instructions for tikoci/rosetta, a project described as: MCP Server with RouterOS docs + commands + products + changelogs, using SQLite-as-RAG, sourced from MikroTik.