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/jmrplens/gitlab-mcp-server/test-goroutinesgit clone --depth 1 https://github.com/jmrplens/gitlab-mcp-serverWrote 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/jmrplens/gitlab-mcp-server/test-goroutines)<a href="https://agentmods.dev/instructions/jmrplens/gitlab-mcp-server/test-goroutines"><img src="https://agentmods.dev/badge/instructions/jmrplens/gitlab-mcp-server/test-goroutines.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 | $0.01098 | $0.01098 |
| Opus 5 | $0.00549 | $0.00549 |
| Sonnet 5 | $0.00220 | $0.00220 |
| Haiku 4.5 | $0.00110 | $0.00110 |
Grade A, and why
gitlab-mcp-server test-goroutines.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 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.
How it starts
The opening of the file, as written. The whole thing — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Assertions off the test goroutine
testing.T.FailNow — and therefore t.Fatal/t.Fatalf — must be called from
the goroutine running the test. Anywhere else (an http.HandlerFunc literal,
a go statement, an errgroup.Group.Go task, an MCP tool handler registered
on an in-memory server) it only terminates that goroutine: the HTTP response
is truncated or never written, the client observes a transport error, and the
test continues against the wreckage. go vet's testinggoroutine analyzer
does not catch handler literals; cmd/audit_test_goroutines does
(make audit-test-goroutines / make check-test-goroutines).
The six rules
Every assertion inside a function literal that crosses a goroutine boundary MUST follow all of these:
- Never
t.Fatal/t.Fatalf/t.FailNowinside the literal. Uset.Errorf(or record and assert later — see rule 5). - Always
returnimmediately after thet.Errorf, even when it is the last statement.t.Fatalprovided that exit for free; a later edit appending code below a baret.Errorfsilently reintroduces the bug (this exact mistake produced a nil-dereference panic in PR #270). - Write a deterministic response before returning. For HTTP handlers,
http.Error(w, "<what failed>", http.StatusInternalServerError); for MCP tool handlers, an error result. Without it the client silently receives a200with an empty body — a worse signal than the EOF it replaces. - Nothing the failed check would have validated may be used afterwards.
If the check guarded a nil (
r.MultipartForm, a decoded struct), thereturnmust come before any use of it. - Prefer recording over asserting. Store observed values in locals (or a struct) inside the handler and assert on the test goroutine after the client call returns. This removes the problem instead of mitigating it.
- "Must not be called" guards become a recorded flag —
var called atomic.Bool(oratomic.Int64for counts) written in the handler and asserted afterwards. Theatomicis mandatory: an unsynchronised variable written from the handler goroutine is a data race.
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.
- 3d ago First seen · 93 lines · 1,098 tokens per session scan A 6f78b6cfb481
gitlab-mcp-server test-goroutines.instructions.md is an instructions file published in the GitHub repository jmrplens/gitlab-mcp-server (31 stars, last pushed 4d ago), licensed MIT. It adds 1,098 tokens to every session, about $0.0055 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
awesome-go AGENTS.md
AGENTS.md instructions for avelino/awesome-go, covering awesome-go · llm contribution guide, project snapshot, when modifying the awesome list, coding guidelines and testing & validation.
go-micro CLAUDE.md
Instructions for micro/go-micro, covering claude.md - go micro project guide, project overview, build & test, run all tests and run tests for a specific package.
go-micro AGENTS.md
Instructions for micro/go-micro, covering repository agent instructions and pull requests from codex tasks.
jwx AGENTS.md
AGENTS.md instructions for lestrrat-go/jwx, covering agents.md, for module consumers, go version, goexperiment and module path vs physical layout.
azure-sdk-for-go go-examples.instructions.md
Instructions for Azure/azure-sdk-for-go, a project described as: This repository is for active development of the Azure SDK for Go. For consumers of the SDK we recommend visiting our public developer docs at.
azure-sdk-for-go copilot-instructions.md
Copilot instructions for Azure/azure-sdk-for-go, covering generated code — do not suggest changes, prerequisites, local sdk generation and package lifecycle (typespec), authoritative reference and default behaviors.