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/webern/mx/corpus-and-testsgit clone --depth 1 https://github.com/webern/mxWhat 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 | $0.00282 | $0.00282 |
| Opus 5 | $0.00141 | $0.00141 |
| Sonnet 5 | $0.00056 | $0.00056 |
| Haiku 4.5 | $0.00028 | $0.00028 |
Grade A, and why
mx corpus-and-tests.instructions.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.
What it actually says
data/ is the MusicXML test corpus; src/private/mxtest/ holds the test suites
(see data/README.md). Review for:
- Adding or removing a corpus file requires: the pinned count bump in
src/private/mxtest/corert/CoreRoundtripTest.cpp, regenerated audit artifacts (make audit: the*.features.xmlsidecar anddata/corpus.xml), and a greenmake core-roundtrip-test. Flag a corpus change missing any of these. - A deliberately unparseable file needs a sibling
<name>.invalidmarker explaining why; a schema-invalid-but-lenient file needs a<name>.fixup.xmlsidecar. New fixups must follow the uniform leniency policy indata/README.md, not invent per-file rules. roundtrip-baseline.txt(the api round-trip pass list) only grows by deliberate commits; flag removals or unexplained churn.ApiLoadSmokeTestonly proves a file imports without crashing; a feature claim needs a fixture pinned in the read->write->read gate (make api-roundtrip).- Test code follows the same C++ rules as the library (no new anonymous namespaces).
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 · 20 lines · 282 tokens per session scan A 74239602a87f
mx corpus-and-tests.instructions.md is an instructions file published in the GitHub repository webern/mx (93 stars, last pushed 4d ago), licensed MIT. It adds 282 tokens to every session, about $0.0014 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 2026-08-30.
Other instructions, from other repositories
cxxmcp AGENTS.md
Instructions for caomengxuan666/cxxmcp, covering agent instructions, source of truth, do not touch casually, public sdk boundary and build options and package contract.
search-engine-core mongodb-bson.instructions.md
MongoDB BSON basic builder and instance initialization.
search-engine-core cpp-api-patterns.instructions.md
Result , BSON, and MongoDB C++ API patterns (zero-error).
search-engine-core code-style.instructions.md
Error handling, file organization, security, performance.
search-engine-core controllers.instructions.md
Controller registration and lazy initialization.
search-engine-core uwebsockets.instructions.md
Instructions for hatefsystems/search-engine-core, covering uwebsockets rules, mandatory: ondata + onaborted and full post template.