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/getsentry/vitest-evals/agents-mdgit clone --depth 1 https://github.com/getsentry/vitest-evalsWhat 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 | $0.00626 | $0.00626 |
| Opus 5 | $0.00313 | $0.00313 |
| Sonnet 5 | $0.00125 | $0.00125 |
| Haiku 4.5 | $0.00063 | $0.00063 |
Grade A, and why
vitest-evals 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 — 46 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Instructions
Package Manager
- Use pnpm:
pnpm install,pnpm test,pnpm typecheck,pnpm build,pnpm evals
Repository Structure
packages/
core/ # normalized session/run types, transcript events, report metadata
vitest-evals/ # root describeEval, judges, matchers, reporter, legacy; re-exports core
harness-ai-sdk/ # AI SDK adapter into HarnessRun
harness-openai-agents/ # OpenAI Agents adapter into HarnessRun
harness-pi-ai/ # pi-ai adapter, wrapped tool runtime, tool replay
github-reporter/ # GitHub Action reporter: collection, summary, annotations
report-ui/ # HTML report UI and its transcript/trace model
docs/ # docs site, generated API reference, llms.txt
apps/ # demo-ai-sdk, demo-openai-agents, demo-pi (realistic demo fixtures)
File-Scoped Commands
| Task | Command |
|---|---|
| Lint file | pnpm exec biome lint path/to/file.ts |
| Format file | pnpm exec biome format --write path/to/file.ts |
| Test file | pnpm exec vitest run path/to/file.test.ts -c vitest.config.ts |
| Eval file | pnpm exec vitest run path/to/file.eval.ts -c vitest.config.ts --reporter=./packages/vitest-evals/src/reporter.ts |
Key Conventions
- Root API work is harness-first and judge-first.
- Legacy scorer-first changes stay under
packages/vitest-evals/src/legacy/.... - Keep normalized session and run data JSON-serializable.
- Keep
UsageSummaryto stable usage units such as tokens, tools, retries, provider, and model. Do not add first-class cost fields; provider-specific cost estimates belong inusage.metadata. - GitHub reporting is action-first: document workflows with
uses: getsentry/vitest-evals@v0, not CLI commands. - Sharded eval reporting uses distinct JSON artifacts per matrix job and one reducer job that publishes the combined action report.
- Root GitHub Action releases keep source and bundled tags separate: Craft publishes the source release,
vX.Y.Z-srcpreserves the source baseline, andvX/vX.Y.Zpoint at the bundled action commit. Keep the Craftgithubtarget filtered away from package artifacts. - Update
README.md,packages/vitest-evals/README.md, and relevant docs when product shape changes. - Follow
policies/code-comments.mdfor exported-function JSDoc and non-obvious comments only.
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 · 46 lines · 626 tokens per session scan A c823b5ec8dee
vitest-evals AGENTS.md is an instructions file published in the GitHub repository getsentry/vitest-evals (341 stars, last pushed 25d ago), licensed Apache-2.0. It adds 626 tokens to every session, about $0.0031 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
plausible-mcp AGENTS.md
Instructions for getsentry/plausible-mcp, covering before calling a change green and reference.
XcodeBuildMCP-iOS-Template copilot-instructions.md
Instructions for getsentry/XcodeBuildMCP-iOS-Template, a project described as: A scaffold template for iOS app projects.
plausible-mcp CLAUDE.md
Instructions for getsentry/plausible-mcp, a project described as: MCP server for Plausible Analytics — query traffic, conversions, and compare periods from any AI tool.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.