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/basic-bit/soundiiz-mcp/agents-mdgit clone --depth 1 https://github.com/BASIC-BIT/soundiiz-mcpWhat 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 | $0.01695 | $0.01695 |
| Opus 5 | $0.00847 | $0.00847 |
| Sonnet 5 | $0.00339 | $0.00339 |
| Haiku 4.5 | $0.00169 | $0.00169 |
Grade A, and why
soundiiz-mcp AGENTS.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 2d 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 — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS
This repo uses linting, typechecking, and tests to validate changes.
Workflow
- After significant code changes (or at the end of a work loop), run
npm run check. - For a single step, use
npm run lint,npm run typecheck, ornpm test. - After making a change, run at least one relevant targeted test (or add/edit one) and confirm it passes before reporting back.
- For PR review workflow: before pushing any new commits, respond to each open review comment (reply or reaction) and resolve the thread.
- Keep stdout reserved for MCP protocol; log to stderr only.
- Config defaults live in
src/config/defaults.json; override viaSOUNDIIZ_MCP_CONFIG_FILE(env overrides still supported). - Writes are enabled by default. Set
writes.allow=falseto lock the server to GET-only mode. - Destructive writes (DELETE only) require a confirmation token unless
writes.confirmDestructive=false.soundiiz_sync_triggeris a non-destructive write and executes in one call. - Regenerate tool catalog docs after spec updates:
npm run generate:tools-docs. - The brand mark lives in
scripts/lib/mark.ts. Do not hand-editassets/logo.svgorassets/social-preview.svg. Edit the mark module, then runnpm run build:assetsto regenerate both surfaces and the social-preview PNG together. The two assets share gradient ids and geometry by construction, so they cannot drift. - Refetch the Soundiiz OpenAPI spec when the upstream BETA changes:
npm run sync:spec. This rewritesspecs/soundiiz-openapi.json. Diff the file before committing — the spec is in BETA and may shift. - Regenerate Zod schemas after spec updates:
npm run generate:schemas(updatessrc/generated/soundiiz-schemas.ts; do not edit manually). - Regenerate mock test schemas after spec tweaks:
npm run generate:test-schemas.
Tool ergonomics goals (distilled)
- Prefer names / shortcodes / titles over numeric IDs when reasonable: tools should accept human inputs as first-class arguments where the API allows.
- Keep tools strict and explicit: each tool has a single, obvious purpose and should not hide search/fallback behaviors internally.
- Fail cleanly with guidance: when a lookup fails, return a structured "not found" response that points to the next obvious tool.
- Chainable outputs: successful responses should include both human labels and IDs for follow-up calls.
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.
- 2d ago First seen · 89 lines · 1,695 tokens per session scan A a59cdc31958e
soundiiz-mcp AGENTS.md is an instructions file published in the GitHub repository BASIC-BIT/soundiiz-mcp (1 stars, last pushed 1mo ago), licensed MIT. It adds 1,695 tokens to every session, about $0.0085 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
PatchWarden AGENTS.md
Instructions for jiezeng2004-design/PatchWarden, covering patchwarden agent rules, commands, safety contracts, changes and release and codex memory.
toon-memory AGENTS.md
Instructions for LuiggiVal08/toon-memory, covering agents.md, build & verify, architecture, smarter memory pattern and visual memory pattern.
vrchat-mcp AGENTS.md
Instructions for BASIC-BIT/vrchat-mcp, covering agents, tool ergonomics goals (distilled), curated output philosophy, specific notes (current direction) and caching considerations (planned).
mcp-qobuz copilot-instructions.md
Copilot instructions for kud/mcp-qobuz, covering copilot instructions — mcp-qobuz, project context, content rules, suppression rules and review format.
fcp-midi CLAUDE.md
Instructions for os-tack/fcp-midi, covering fcp-midi, project overview, architecture, key directories and commands.
roon-mcp AGENTS.md
AGENTS.md instructions for bestimmaa/roon-mcp, covering roon mcp — project notes and references.