ref: Fork orchestrion-js, import-in-the-middle and require-in-the-middle

.changeset/hot-boats-fail.md

@@ -0,0 +1,5 @@
+---
+"braintrust": patch
+---
+
+ref: Fork `orchestrion-js`, `import-in-the-middle` and `require-in-the-middle`

AGENTS.md

@@ -71,3 +71,11 @@ pnpm run lint            # Run eslint checks
 pnpm run fix:formatting  # Auto-fix formatting
 pnpm run fix:lint        # Auto-fix eslint issues
 ```
+
+## Vendored Forks and Licenses
+
+When touching forked third-party code in this repository, including the
+vendored `import-in-the-middle`, `require-in-the-middle`, or `orchestrion-js`
+code, preserve and respect the upstream license requirements. Keep copyright
+notices, license files, provenance notes, and `js/NOTICE` entries accurate when
+copying, updating, or materially modifying vendored code.

e2e/scenarios/pi-coding-agent-instrumentation/__snapshots__/pi-coding-agent-v079-auto-hook.span-tree.json

@@ -17,7 +17,7 @@
                   "children": [],
                   "input": [
                     {
-                      "content": "You are an expert coding assistant operating inside pi, a coding agent harness. You help users by reading files, executing commands, editing code, and writing new files.\n\nAvailable tools:\n- bash: Execute bash commands (ls, grep, find, etc.)\n\nIn addition to the tools above, you may have access to other custom tools depending on the project.\n\nGuidelines:\n- Use bash for file operations like ls, rg, find\n- Be concise in your responses\n- Show file paths clearly when working with files\n\nPi documentation (read only when the user asks about pi itself, its SDK, extensions, themes, skills, or TUI):\n- Main documentation: <repo>/e2e/.bt-tmp/scenario-deps/pi-coding-agent-instrumentation-locked-<hash>/node_modules/.pnpm/@earendil-works+pi-coding-agent@0.79.1_ws@8.21.0_zod@4.4.3/node_modules/@earendil-works/pi-coding-agent/README.md\n- Additional docs: <repo>/e2e/.bt-tmp/scenario-deps/pi-coding-agent-instrumentation-locked-<hash>/node_modules/.pnpm/@earendil-works+pi-coding-agent@0.79.1_ws@8.21.0_zod@4.4.3/node_modules/@earendil-works/pi-coding-agent/docs\n- Examples: <repo>/e2e/.bt-tmp/scenario-deps/pi-coding-agent-instrumentation-locked-<hash>/node_modules/.pnpm/@earendil-works+pi-coding-agent@0.79.1_ws@8.21.0_zod@4.4.3/node_modules/@earendil-works/pi-coding-agent/examples (extensions, custom tools, SDK)\n- When reading pi docs or examples, resolve docs/... under Additional docs and examples/... under Examples, not the current working directory\n- When asked about: extensions (docs/extensions.md, examples/extensions/), themes (docs/themes.md), skills (docs/skills.md), prompt templates (docs/prompt-templates.md), TUI components (docs/tui.md), keybindings (docs/keybindings.md), SDK integrations (docs/sdk.md), custom providers (docs/custom-provider.md), adding models (docs/models.md), pi packages (docs/packages.md)\n- When working on pi topics, read the docs and examples, and follow .md cross-references before implementing\n- Always read pi .md files completely and follow links to related docs (e.g., tui.md for TUI API details)\n\n<project_context>\n\nProject-specific instructions and guidelines:\n\n<project_instructions path=\"<repo>/AGENTS.md\">\n# Braintrust JavaScript SDK Monorepo\n\nTypeScript SDKs and integrations for Braintrust. Uses `pnpm` workspaces.\n\n## Repository Structure\n\n```text\n.\n├── js/             # Main `braintrust` package\n├── integrations/   # Integration packages (@braintrust/*)\n├── e2e/            # End-to-end scenario tests (mock server + subprocess isolation)\n├── docs/           # Docs and reference material\n└── internal/       # Internal test fixtures and golden projects\n```\n\n## Setup\n\n```bash\nmise install        # Install toolchain and dependencies\n```\n\n## Build\n\n```bash\npnpm run build      # Build all workspace packages (from repo root)\n```\n\n## Instrumentation\n\nUse the normal Orchestrion config plus plugin/channel path by default. Special-case source patches should be rare exceptions only when the target SDK cannot be instrumented through the standard transformer path, and the reason should be documented next to the patch.\n\n## Testing\n\nUses Vitest. Prefer running the **narrowest relevant test** rather than the full suite.\n\n**From `js/` directory:**\n\n```bash\npnpm test                         # Core vitest suite (excludes wrappers)\npnpm test -- -t \"test name\"       # Filter by test name\npnpm run test:checks              # Hermetic tests (core + vitest wrapper)\n```\n\n**E2E tests (`e2e/`):**\n\nEach scenario runs the SDK in a subprocess against a mock Braintrust server and snapshots the results. No API keys required for replay; recording needs provider keys.\n\n```bash\npnpm run test:e2e                 # Run all e2e scenarios (from repo root)\npnpm run test:e2e:update          # Update e2e snapshots without re-recording cassettes\npnpm run test:e2e:record          # Re-record provider cassettes and update snapshots\n```\n\nWhen adding or modifying e2e tests, run the relevant e2e verification twice before stopping so flakes are caught proactively. After running `pnpm run test:e2e:update` or `pnpm run test:e2e:record`, always run the normal e2e tests afterward to verify there is no snapshot drift or unstable output.\n\nSpan-tree snapshots are paired: `*.span-tree.json` is the structural contract, and `*.span-tree.txt` is the human-readable ASCII tree generated from the same normalized spans. Both files are asserted and should be updated together through `pnpm run test:e2e:update` or `pnpm run test:e2e:record`; do not hand-edit only one side of the pair.\n\n**From repo root:**\n\n```bash\npnpm run test       # Run all workspace tests via turbo\n```\n\n## Linting & Formatting\n\nRun from the repo root. **Always run `fix:formatting` before committing** — there is a pre-commit hook that will reject unformatted code.\n\n```bash\npnpm run formatting      # Check formatting (prettier)\npnpm run lint            # Run eslint checks\npnpm run fix:formatting  # Auto-fix formatting\npnpm run fix:lint        # Auto-fix eslint issues\n```\n\n</project_instructions>\n\n</project_context>\n\nCurrent date: <timestamp>\nCurrent working directory: <repo>/e2e/scenarios/pi-coding-agent-instrumentation/",
+                      "content": "You are an expert coding assistant operating inside pi, a coding agent harness. You help users by reading files, executing commands, editing code, and writing new files.\n\nAvailable tools:\n- bash: Execute bash commands (ls, grep, find, etc.)\n\nIn addition to the tools above, you may have access to other custom tools depending on the project.\n\nGuidelines:\n- Use bash for file operations like ls, rg, find\n- Be concise in your responses\n- Show file paths clearly when working with files\n\nPi documentation (read only when the user asks about pi itself, its SDK, extensions, themes, skills, or TUI):\n- Main documentation: <repo>/e2e/.bt-tmp/scenario-deps/pi-coding-agent-instrumentation-locked-<hash>/node_modules/.pnpm/@earendil-works+pi-coding-agent@0.79.1_ws@8.21.0_zod@4.4.3/node_modules/@earendil-works/pi-coding-agent/README.md\n- Additional docs: <repo>/e2e/.bt-tmp/scenario-deps/pi-coding-agent-instrumentation-locked-<hash>/node_modules/.pnpm/@earendil-works+pi-coding-agent@0.79.1_ws@8.21.0_zod@4.4.3/node_modules/@earendil-works/pi-coding-agent/docs\n- Examples: <repo>/e2e/.bt-tmp/scenario-deps/pi-coding-agent-instrumentation-locked-<hash>/node_modules/.pnpm/@earendil-works+pi-coding-agent@0.79.1_ws@8.21.0_zod@4.4.3/node_modules/@earendil-works/pi-coding-agent/examples (extensions, custom tools, SDK)\n- When reading pi docs or examples, resolve docs/... under Additional docs and examples/... under Examples, not the current working directory\n- When asked about: extensions (docs/extensions.md, examples/extensions/), themes (docs/themes.md), skills (docs/skills.md), prompt templates (docs/prompt-templates.md), TUI components (docs/tui.md), keybindings (docs/keybindings.md), SDK integrations (docs/sdk.md), custom providers (docs/custom-provider.md), adding models (docs/models.md), pi packages (docs/packages.md)\n- When working on pi topics, read the docs and examples, and follow .md cross-references before implementing\n- Always read pi .md files completely and follow links to related docs (e.g., tui.md for TUI API details)\n\n<project_context>\n\nProject-specific instructions and guidelines:\n\n<project_instructions path=\"<repo>/AGENTS.md\">\n# Braintrust JavaScript SDK Monorepo\n\nTypeScript SDKs and integrations for Braintrust. Uses `pnpm` workspaces.\n\n## Repository Structure\n\n```text\n.\n├── js/             # Main `braintrust` package\n├── integrations/   # Integration packages (@braintrust/*)\n├── e2e/            # End-to-end scenario tests (mock server + subprocess isolation)\n├── docs/           # Docs and reference material\n└── internal/       # Internal test fixtures and golden projects\n```\n\n## Setup\n\n```bash\nmise install        # Install toolchain and dependencies\n```\n\n## Build\n\n```bash\npnpm run build      # Build all workspace packages (from repo root)\n```\n\n## Instrumentation\n\nUse the normal Orchestrion config plus plugin/channel path by default. Special-case source patches should be rare exceptions only when the target SDK cannot be instrumented through the standard transformer path, and the reason should be documented next to the patch.\n\n## Testing\n\nUses Vitest. Prefer running the **narrowest relevant test** rather than the full suite.\n\n**From `js/` directory:**\n\n```bash\npnpm test                         # Core vitest suite (excludes wrappers)\npnpm test -- -t \"test name\"       # Filter by test name\npnpm run test:checks              # Hermetic tests (core + vitest wrapper)\n```\n\n**E2E tests (`e2e/`):**\n\nEach scenario runs the SDK in a subprocess against a mock Braintrust server and snapshots the results. No API keys required for replay; recording needs provider keys.\n\n```bash\npnpm run test:e2e                 # Run all e2e scenarios (from repo root)\npnpm run test:e2e:update          # Update e2e snapshots without re-recording cassettes\npnpm run test:e2e:record          # Re-record provider cassettes and update snapshots\n```\n\nWhen adding or modifying e2e tests, run the relevant e2e verification twice before stopping so flakes are caught proactively. After running `pnpm run test:e2e:update` or `pnpm run test:e2e:record`, always run the normal e2e tests afterward to verify there is no snapshot drift or unstable output.\n\nSpan-tree snapshots are paired: `*.span-tree.json` is the structural contract, and `*.span-tree.txt` is the human-readable ASCII tree generated from the same normalized spans. Both files are asserted and should be updated together through `pnpm run test:e2e:update` or `pnpm run test:e2e:record`; do not hand-edit only one side of the pair.\n\n**From repo root:**\n\n```bash\npnpm run test       # Run all workspace tests via turbo\n```\n\n## Linting & Formatting\n\nRun from the repo root. **Always run `fix:formatting` before committing** — there is a pre-commit hook that will reject unformatted code.\n\n```bash\npnpm run formatting      # Check formatting (prettier)\npnpm run lint            # Run eslint checks\npnpm run fix:formatting  # Auto-fix formatting\npnpm run fix:lint        # Auto-fix eslint issues\n```\n\n## Vendored Forks and Licenses\n\nWhen touching forked third-party code in this repository, including the\nvendored `import-in-the-middle`, `require-in-the-middle`, or `orchestrion-js`\ncode, preserve and respect the upstream license requirements. Keep copyright\nnotices, license files, provenance notes, and `js/NOTICE` entries accurate when\ncopying, updating, or materially modifying vendored code.\n\n</project_instructions>\n\n</project_context>\n\nCurrent date: <timestamp>\nCurrent working directory: <repo>/e2e/scenarios/pi-coding-agent-instrumentation/",
                       "role": "system"
                     },
                     {