Add Devin skills for testing, setup, codegen, CI, releases, templates, and triage

.agents/skills/testing-cli/SKILL.md

@@ -0,0 +1,47 @@
+---
+name: testing-cli
+description: "Build, run, and test the E2B CLI (packages/cli) locally. Use when changing CLI commands, running its vitest suites, or executing the CLI against real sandboxes."
+---
+
+# Testing the CLI
+
+All commands run in `packages/cli`. Use Node 24 (`nvm use 24`) and pnpm.
+
+## Build and run locally
+
+```bash
+pnpm build                       # tsc typecheck + tsdown bundle -> dist/index.js
+node dist/index.js --help
+node dist/index.js sandbox list
+```
+
+Auth: the CLI reads `E2B_API_KEY` from the environment first, then `~/.e2b/config.json` (or `.env.local` at the repo root). Never run `e2b auth login` in headless environments — export the key instead.
+
+Useful non-interactive patterns:
+
+```bash
+node dist/index.js sandbox create base --detach          # returns sandbox ID, no attached terminal
+node dist/index.js sandbox exec <id> -- bash -lc 'pwd'   # `--` stops CLI flag parsing
+node dist/index.js sandbox kill <id>
+```
+
+## Automated tests
+
+```bash
+pnpm run test                          # vitest; globalSetup runs `pnpm build` first
+npx vitest run tests/utils/table.test.ts   # single file
+```
+
+- Tests spawn the **built** CLI (`dist/index.js`) via helpers in `tests/setup.ts` (`runCli`, `runCliWithPipedStdin`) — rebuild happens automatically through globalSetup, but if you bypass vitest, run `pnpm build` yourself after editing `src/`.
+- `tests/commands/**` cover command behavior; some hit the real API and need `E2B_API_KEY`.
+- Unit tests import from `src/` directly (vitest aliases `e2b` to `../js-sdk/src`), so keep command logic in exported, testable functions (see `buildTableRows`/`sortSandboxes` in `src/commands/sandbox/list.ts`).
+
+For checking rendered output (tables, colors, alignment), use the `verifying-cli-visual-output` skill.
+
+## Before committing
+
+```bash
+pnpm run lint && pnpm run typecheck
+```
+
+Public-surface changes need a changeset (`pnpm changeset` at repo root).

.agents/skills/testing-js-sdk/SKILL.md

@@ -0,0 +1,50 @@
+---
+name: testing-js-sdk
+description: "Run and write tests for the E2B JavaScript SDK (packages/js-sdk). Use when changing js-sdk code, debugging its vitest suites, or verifying SDK behavior against real sandboxes."
+---
+
+# Testing the JS SDK
+
+All commands run in `packages/js-sdk`. Use Node 24 (`nvm use 24`) and pnpm.
+
+## Test projects
+
+Tests are organized into vitest projects in `vitest.config.mts`:
+
+- `unit` — everything in `tests/**/*.test.ts` except runtimes, template, and connectionConfig. Many of these are **integration tests that create real sandboxes** and require `E2B_API_KEY`.
+- `template` — template builder tests (`tests/template/**`), 180s timeout, require `E2B_API_KEY`.
+- `connectionConfig` — offline config tests.
+- `browser` — Playwright/chromium tests (`pnpm run playwright:install` first).
+
+## Running
+
+```bash
+# everything (needs E2B_API_KEY)
+pnpm run test
+
+# one file — fastest loop, preferred while iterating
+npx vitest run tests/api/inflight.test.ts
+
+# one project
+npx vitest run --project connectionConfig
+
+# alternate runtimes
+pnpm run test:bun    # bun: unit + connectionConfig + template
+pnpm run test:deno   # deno: same projects
+```
+
+`E2B_API_KEY` is read from the environment or from `.env` via dotenv (the repo also keeps defaults in `.env.local` at the root or `~/.e2b/config.json`). Purely offline unit tests (e.g. `tests/api/inflight.test.ts`, `tests/utils.test.ts`) run without a key.
+
+## Writing tests
+
+- Use the helpers in `tests/setup.ts`: `sandboxTest` / `templateTest` fixtures create and clean up sandboxes; `isDebug` gates behavior when `E2B_DEBUG` is set (local envd at debug port).
+- Offline tests that need HTTP mock the API with `msw`; see `tests/api/` for patterns. Test isolation is required (`isolate: true`) because suites patch global fetch.
+- SDK changes must be mirrored in the Python SDK (sync + async) with equivalent tests — see `testing-python-sdk`.
+
+## Before committing
+
+```bash
+pnpm run lint && pnpm run typecheck
+```
+
+Public-surface changes need a changeset (`pnpm changeset` at repo root).

.agents/skills/testing-python-sdk/SKILL.md

@@ -0,0 +1,44 @@
+---
+name: testing-python-sdk
+description: "Run and write tests for the E2B Python SDK (packages/python-sdk). Use when changing python-sdk code (sync or async), running pytest suites, or mirroring JS SDK changes in Python."
+---
+
+# Testing the Python SDK
+
+All commands run in `packages/python-sdk`. Use uv for everything (`uv run ...`); never pip.
+
+## Layout
+
+- `tests/sync/` and `tests/async/` — integration tests against real sandboxes (need `E2B_API_KEY`). Sync and async variants must stay equivalent.
+- `tests/test_*.py` (top level) — offline unit tests (transports, codecs, config parsing, etc.).
+- `tests/conftest.py` — fixtures that create/clean up sandboxes; `pytest.ini` sets `asyncio_mode=auto`, a 30s per-test timeout, and `pythonpath = tests` for shared helpers like `envd_frame_server`.
+
+## Running
+
+```bash
+# full suite, 4 workers (needs E2B_API_KEY in env)
+pnpm run test            # == uv run pytest -n 4 --verbose -x
+
+# single file / test — preferred while iterating
+uv run pytest tests/test_paginator.py -v
+uv run pytest tests/sync/sandbox_sync/test_create.py -v -k "metadata"
+
+# offline-only quick check (skip integration dirs)
+uv run pytest tests --ignore=tests/sync --ignore=tests/async -q
+```
+
+The `skip_debug` marker skips a test when `E2B_DEBUG` is set (local envd).
+
+## Writing tests
+
+- Every behavior change must land in **both** sync and async implementations with matching tests in `tests/sync/` and `tests/async/` (and mirror the JS SDK — see `testing-js-sdk`).
+- Async tests need no decorator (`asyncio_mode=auto`).
+- Reuse conftest fixtures rather than creating sandboxes by hand.
+
+## Before committing
+
+```bash
+pnpm run lint && pnpm run typecheck   # ruff check/format + ty check
+```
+
+Public-surface changes need a changeset (`pnpm changeset` at repo root).

.agents/skills/verifying-cli-visual-output/SKILL.md

@@ -0,0 +1,51 @@
+---
+name: verifying-cli-visual-output
+description: "Verify the visual/terminal output of the E2B CLI: tables, colors, alignment, spinners, TTY vs piped behavior. Use when changing anything the CLI prints or when reviewing output formatting."
+---
+
+# Verifying CLI Visual Output
+
+The CLI's output style follows kubectl: borderless tables, uppercase headers, left-aligned columns with 3-space padding, no trailing whitespace (`packages/cli/src/utils/table.ts` → `renderTable`). Colors/bold come from `chalk` (`src/utils/format.ts`).
+
+## 1. Unit-test the rendering (preferred)
+
+Capture `console.log` lines and assert exact strings — see `tests/utils/table.test.ts`:
+
+```ts
+const lines: string[] = []
+vi.spyOn(console, 'log').mockImplementation((l: string) => lines.push(l))
+renderTable(rows, columns)
+expect(lines).toEqual(['SANDBOX ID   NAME', 'sbx-1        alpha'])
+```
+
+Keep row-building logic in exported pure functions (e.g. `buildTableRows` in `src/commands/sandbox/list.ts`) so formatting is testable without the API.
+
+## 2. Eyeball the real output
+
+```bash
+cd packages/cli && pnpm build
+node dist/index.js sandbox list
+```
+
+Checks to make by eye:
+- Headers uppercase; columns aligned even with wide cells (widths use `wcswidth`, so CJK/emoji count as 2 cells).
+- No borders, no trailing whitespace (`node dist/index.js sandbox list | cat -A` — no spaces before `$`).
+- Long values (metadata JSON) don't break alignment of preceding columns.
+- Empty result sets print a sensible message, not a lone header or a crash.
+
+## 3. TTY vs piped behavior
+
+chalk auto-strips colors when stdout is not a TTY, so piped output must stay clean and parseable:
+
+```bash
+node dist/index.js sandbox list | head          # no ANSI escape codes expected
+node dist/index.js sandbox list | grep -c $'\e' # should be 0
+FORCE_COLOR=1 node dist/index.js sandbox list   # force colors while piping, to inspect them
+script -qec "node dist/index.js sandbox list" /dev/null  # run under a real PTY
+```
+
+Interactive commands (spinners, prompts via `inquirer`) need a PTY — use the `script` trick above or a tty-enabled shell; never leave them attached in CI-style runs (prefer `--detach` variants).
+
+## 4. Screenshot for PRs
+
+For user-facing output changes, run the command in a real terminal, take a screenshot, and embed it in the PR description — reviewers care about how it looks, not just the strings.