When SSH disconnects with exit code 255, the server is still running.
Previously the warn message ("SSH connection lost") was followed by the
full reportScriptFailure block, which was contradictory. Now we return
undefined after the warn so reportScriptFailure is skipped entirely.
Fixes#2185
Agent: issue-fixer
Co-authored-by: B <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
When a user ran `spawn claude --dry-run`, the dry-run flag was silently
ignored and a real server was provisioned. `cmdAgentInteractive` was
passing `dryRun` in the `debug` parameter position of `execScript`, so
no preview was shown and `SPAWN_DEBUG=1` was set instead.
Fix:
- Export `showDryRunPreview` from `run.ts`
- Import and call it in `cmdAgentInteractive` after cloud selection
- Return early when `dryRun` is set (matches `cmdRun` behaviour)
- Pass `undefined` for the `debug` argument (interactive path has no
debug flag)
Agent: code-health
Co-authored-by: B <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
Previously, saveVmConnection wrote to a single last-connection.json temp
file that was only merged into history.json lazily when spawn ls was run.
This caused connections to be silently dropped when:
- Two servers spawned before running spawn ls (file overwritten)
- The last history record already had a connection (merge skipped)
Now saveVmConnection writes directly into history.json by finding the
most recent record matching the cloud with no connection yet. The temp
file is still written for backward compatibility but is no longer the
primary storage.
Also fixes saveLaunchCmd to update history.json directly, and
consolidates sprite's local saveVmConnection to use the shared one.
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: A <258483684+la14-1@users.noreply.github.com>
- Update key-request.sh comment that referenced non-existent
loadTokenFromConfig function in digitalocean.ts
- Update test comments referencing validateAgent/validateCloud
which were renamed to validateEntity
Co-authored-by: spawn-qa-bot <qa@openrouter.ai>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
The auto-update path in update-check.ts and the manual `spawn update` command
in commands/update.ts were missing --proto '=https' on their curl calls that
download and execute the install script. Without it, curl may follow redirects
to non-HTTPS URLs on hostile networks (MITM/DNS hijacking).
- update-check.ts: add --proto =https to execFileSync curl args
- commands/update.ts: replace execSync shell pipe with safe two-step
execFileSync pattern (fetch script via curl --proto =https, then
execute via bash -c) — matches the pattern already in update-check.ts
Same vulnerability class as PR #2172 (TypeScript files) and PR #2160 (shell
scripts); those PRs missed these two code paths.
Agent: security-auditor
Co-authored-by: B <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
Add a 5th teammate (record-keeper) to the QA quality cycle that keeps
README.md in sync with source-of-truth files. Uses a conservative
three-gate check (matrix drift, commands drift, troubleshooting gaps)
and only makes changes when drift is detected. Includes safeguards:
30-line diff limit, prohibited sections list, and source citations
required in PR body.
Co-authored-by: spawn-bot <spawn-bot@openrouter.ai>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* fix(security): add --proto '=https' to curl calls in TypeScript provisioning
Fixes#2169
Agent: security-auditor
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
* fix(lint): break long lines for biome format compliance
Agent: pr-maintainer
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
---------
Co-authored-by: B <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
Replace the Nous Research org avatar with the actual Staff of Hermes
(⚕) symbol from the hermes-agent page favicon. Sourced from the
WordPress emoji SVG and converted to 180x180 PNG.
Co-authored-by: lab <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Worktrees don't share node_modules with the main checkout. Without
`bun install`, tests and biome fail with "Cannot find package" errors
that block the pre-merge hook.
Co-authored-by: lab <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Renames create_server, delete_server, ssh_keys, and server_types
fixture JSON files to kebab-case for consistency with codebase
conventions. Updates _metadata.json keys and qa-fixtures-prompt
naming convention accordingly.
Co-authored-by: lab <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* feat: migrate shell script URLs to openrouter.ai/labs/spawn CDN
Users on older CLI versions can't auto-update because the repo was restructured
(cli/ → packages/cli/), so old version-check URLs 404. This decouples the CLI
from the repo's internal directory structure:
- Shell script URLs (install, agent scripts, github-auth) now use
openrouter.ai/labs/spawn/* as primary with GitHub raw as fallback
- Version checks now use GitHub release artifact (cli-latest/version)
as primary — a static URL that never changes regardless of repo layout
- CI workflow updated to publish a `version` file alongside cli.js
- Remove GITHUB_RAW_URL_PATTERN validation (no longer needed since
install URL is now a hardcoded CDN string, not interpolated)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* style: fix biome formatting in update-check test
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: CLAUDE.md says biome lint but should say biome check
biome lint only checks lint rules, not formatting. biome check does both.
The hooks and CI already run biome check — the docs were out of sync.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(hooks): PostToolUse hook wasn't running biome on CLI source files
Two bugs in validate-file.ts:
1. Config search only checked 1-2 levels up from the edited file, but
biome.json is at packages/cli/ — 3 levels above src/__tests__/*.ts.
Fix: walk up directories until biome.json is found (or hit root).
2. Ran `biome format` (prints formatted output, always exits 0) instead
of `biome format --check` (exits non-zero if file needs formatting).
Fix: use `biome check` which does lint + format check in one pass.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Adds a PreToolUse hook for Bash that intercepts `gh pr merge` and
`gh pr ready` commands and runs `biome check src/` + `bun test` before
allowing them. Blocks the command if either check fails.
The hook finds the worktree from the command path or falls back to
git rev-parse --show-toplevel.
Co-authored-by: lab <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* refactor: extract inline hook commands to TypeScript scripts in .claude/scripts/
Replace long inline `bash -c '...'` one-liners in .claude/settings.json with
standalone TypeScript scripts that are easier to read, debug, and maintain:
- enforce-worktree.ts: PreToolUse hook ensuring edits happen in worktrees
- validate-file.ts: PostToolUse hook for .sh/.ts file validation
- pre-merge-check.ts: PreToolUse hook running biome + tests before merge
Add .claude/scripts as a bun workspace package (@spawn/hooks).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* refactor: replace manual typeguards with valibot schemas in hook scripts
- Extract shared schemas (FilePathInput, CommandInput, parseStdin) to schemas.ts
- Replace inline multi-level typeof/in checks with v.safeParse() calls
- Add valibot dependency to @spawn/hooks package
- Add CLAUDE.md rule: always prefer valibot over manual typeguards, share schemas
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* refactor: split CLAUDE.md into modular .claude/rules/ files
Split the 437-line monolithic CLAUDE.md into a lean 89-line project overview
plus 9 focused rules files in .claude/rules/ (auto-loaded by Claude Code):
- culture.md — embrace bold changes, parallelize, verify exhaustively
- shell-scripts.md — curl|bash compat, macOS bash 3.x, ESM only, bun not python
- type-safety.md — no `as` assertions, ALWAYS use valibot (never manual typeguards)
- testing.md — bun:test only, no vitest, no subprocess spawning
- git-workflow.md — worktree-first mandatory workflow
- autonomous-loops.md — discovery/refactor service architecture
- discovery.md — how to fill matrix gaps, add clouds/agents
- documentation.md — never commit docs, use .docs/
- cli-version.md — bump version on every CLI change
The type-safety rule now explicitly mandates valibot schemas over manual
typeguard chains in all cases beyond single-primitive narrowing.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(lint): run biome check across all packages in CI
The lint workflow only checked packages/cli/src/. Now it checks all
TypeScript locations in a single biome check command:
- packages/cli/src/ (with GritQL plugins)
- packages/shared/src/ (new biome.json)
- .claude/scripts/ (new biome.json)
- .claude/skills/setup-spa/
Fixed all pre-existing lint/format errors:
- node: protocol on all Node.js built-in imports in hook scripts
- useBlockStatements in packages/shared/src/type-guards.ts
- expand formatting in .claude/skills/setup-spa/main.ts and spa.test.ts
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: lab <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: restore hyphens/underscores in OAuth code regex + add test (#2116)
PR #2116 broke OAuth by restricting the auth code regex to alphanumeric
only. OAuth providers (GitHub, Google, etc.) use hyphens and underscores
in their auth codes, so the stricter regex rejected valid codes.
Changes:
- Extract OAUTH_CODE_REGEX as an exported constant from oauth.ts
- Restore `_-` in the character class: [a-zA-Z0-9_-]{16,128}
- Add oauth-code-validation.test.ts with 20 tests covering:
- Real-world provider formats (hyphens, underscores, mixed)
- Length bounds (16–128)
- Injection prevention (shell, XSS, path traversal, null bytes)
- Character class completeness (explicit regression test for #2116)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* ci: retrigger checks
* fix: restore hyphens in OAuth code regex + add regression test
PR #2116 broke OAuth by restricting the auth code regex to alphanumeric
only. OAuth providers (GitHub, Google) use hyphens and underscores in
their auth codes, so the stricter regex rejected valid codes.
- Extract OAUTH_CODE_REGEX to oauth-constants.ts (zero-dep, testable)
- Restore `_-` in character class: [a-zA-Z0-9_-]{16,128}
- Add regression test covering valid formats, length bounds, injection
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---------
Co-authored-by: spawn-bot <spawn-bot@openrouter.ai>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* fix(history): smart trimming evicts deleted records first, archives overflow
When history exceeds 100 entries, deleted records (useless for `spawn ls`)
are now evicted first. If still over the limit, oldest non-deleted records
are also trimmed. All evicted records are archived to dated backup files
(history-YYYY-MM-DD.json) so nothing is permanently lost.
Previously, blind .slice() could silently discard records with active
connections that `spawn ls` depends on.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* style: fix biome formatting issues
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: lab <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The resolveAndLog via cmdRun describe block in commands-swap-resolve.test.ts
(~113 lines, 5 tests) duplicated display-name resolution coverage already
provided by commands-resolve-run.test.ts. Both files tested case-insensitive
key resolution (CLAUDE->claude, HETZNER->hetzner) and display name resolution
(Codex->codex, Sprite->sprite) on the same code path.
Removed the entire duplicate resolveAndLog section. The detectAndFixSwappedArgs
and prompt-handling-with-swapped-args sections remain, as those test distinct
behavior not covered elsewhere.
-- qa/dedup-scanner
Co-authored-by: spawn-qa-bot <qa@openrouter.ai>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-authored-by: L <6723574+louisgv@users.noreply.github.com>
The block comment in run-path-credential-display.test.ts listed five
functions it claimed to test, but the file only tests two:
- prioritizeCloudsByCredentials
- isRetryableExitCode
Functions buildCredentialStatusLines, formatAuthVarLine, validateRunSecurity,
and validateEntities were never imported or exercised in this file. Removed
the misleading entries so the comment accurately reflects test coverage.
-- qa/code-quality
Co-authored-by: spawn-qa-bot <qa@openrouter.ai>
When the TCP probe succeeds on the final attempt, `attempt` equals
`maxAttempts` after the loop increments it. The previous guard
`attempt >= maxAttempts` then incorrectly threw a timeout error even
though the port was open.
Fix by tracking TCP success with a `tcpOpen` boolean flag and checking
that instead of the attempt counter.
Fixes#2155
Agent: issue-fixer
Co-authored-by: B <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
* refactor: Remove redundant loadTokenFromConfig wrappers in hetzner, daytona, digitalocean
The previous PR (#2151) introduced shared loadApiToken() in shared/ui.ts and
updated hetzner/daytona to delegate to it via thin wrapper functions. This
commit removes the now-unnecessary wrapper functions entirely, inlining the
loadApiToken() calls directly at the callsite.
Also removes the 16-line duplicate loadTokenFromConfig() implementation in
digitalocean.ts (which replicates the same api_key/token field reading and
regex validation logic as loadApiToken) and replaces it with a direct call to
loadApiToken("digitalocean").
-- qa/code-quality
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* bump version to 0.12.12 (main already has 0.12.11)
---------
Co-authored-by: spawn-qa-bot <qa@openrouter.ai>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-authored-by: L <6723574+louisgv@users.noreply.github.com>
The 'promptBundle should skip prompt without --custom' test expected
promptBundle() to return immediately when SPAWN_CUSTOM is unset. But
promptBundle() has no SPAWN_CUSTOM guard — it always shows an interactive
selection prompt unless LIGHTSAIL_BUNDLE or SPAWN_NON_INTERACTIVE=1 is set.
Without SPAWN_NON_INTERACTIVE=1, the test blocks on stdin input and hits
the 5-second bun:test timeout. When run in the full test suite it
appeared to pass due to module import caching from previous tests, making
it a flaky, non-deterministic test.
Remove the test entirely since it tests non-existent behavior.
Co-authored-by: spawn-qa-bot <qa@openrouter.ai>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-authored-by: L <6723574+louisgv@users.noreply.github.com>
Fixes#2156
The spinner was stopped with a success message before the HTTP response
body stream was fully consumed. If the stream failed mid-transfer (network
drop, truncation), users saw "Script downloaded" followed by a confusing
downstream error. Now both the primary and fallback paths await res.text()
before calling s.stop().
Agent: issue-fixer
Co-authored-by: B <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
- Extract duplicate loadTokenFromConfig helper (hetzner + daytona) into
shared loadApiToken() in shared/ui.ts, eliminating 24 lines of
duplicate validation logic across two cloud modules
- Move misplaced FETCH_TIMEOUT and UPDATE_BACKOFF_MS constants in
update-check.ts from the Schemas section into the Constants section
where they belong (stale empty section header fix)
Co-authored-by: spawn-qa-bot <qa@openrouter.ai>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(security): validate env values in cloud_headless_env parser
Reject values containing shell metacharacters ($, backtick, ;, &, |, <, >)
to prevent potential command injection if a cloud driver returns malicious output.
Fixes#2139
Agent: security-auditor
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(security): replace env value blacklist with whitelist regex
The blacklist approach missed dangerous characters like (), quotes,
backslash, newlines, {}, and !. Switch to a whitelist that only allows
[A-Za-z0-9@%+=:,./_-] — a strict safe set sufficient for env values.
Agent: security-auditor
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
---------
Co-authored-by: B <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
The "install commands should be strings" test in the "Agent launch
command consistency" describe block was a full duplicate of the
per-agent "install should be a non-empty string" test already
present in the "Agent required field types" loop.
Co-authored-by: spawn-qa-bot <qa@openrouter.ai>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-authored-by: L <6723574+louisgv@users.noreply.github.com>
The fallback .spawnrc construction (used when provision times out before
.spawnrc is written) had two bugs:
1. zeroclaw case wrongly included OPENAI_API_KEY and OPENAI_BASE_URL —
these are hermes env vars, not zeroclaw's. zeroclaw only needs
ZEROCLAW_PROVIDER=openrouter (plus the base OPENROUTER_API_KEY).
2. hermes and kilocode were missing from the case statement entirely.
- hermes needs OPENAI_BASE_URL and OPENAI_API_KEY (verify_hermes
checks for OPENAI_BASE_URL in .spawnrc)
- kilocode needs KILO_PROVIDER_TYPE=openrouter and
KILO_OPEN_ROUTER_API_KEY (verify_kilocode checks KILO_PROVIDER_TYPE)
Without these fixes, hermes and kilocode would fail verification whenever
provisioning timed out before the normal .spawnrc was written.
Co-authored-by: spawn-qa-bot <qa@openrouter.ai>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
When credentials expire during server deletion, the spinner was running
simultaneously with interactive credential prompts, creating confusing
overlapping UI. Extract ensureDeleteCredentials() to run all credential
checks (which may prompt the user) before starting the deletion spinner.
All 6 cloud providers are covered: AWS, Hetzner, DigitalOcean, GCP,
Daytona, and Sprite.
Fixes#2141
Co-authored-by: B <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Distinguish between 'no local credentials' and 'using spawn-cached credentials'
so users understand why an initial failure message is followed by a success.
Fixes#2142
Agent: ux-engineer
Co-authored-by: B <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(security): add --proto '=https' to all curl bun installer calls
Fixes#2134
All _ensure_bun() functions across aws, hetzner, gcp, local, daytona,
and sprite scripts now enforce HTTPS-only downloads via --proto '=https'.
This prevents MITM attacks during bun installation on remote VMs.
DigitalOcean scripts were already correct and are not changed.
Agent: security-auditor
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(security): add --proto '=https' to bun installer in TS files
Address security reviewer feedback: the same MITM vulnerability
existed in 5 TypeScript programmatic provisioning files.
Agent: pr-maintainer
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(security): quote --proto '=https' in su -c curl calls
The aws.ts and gcp.ts files had --proto =https without quotes inside
su -c '...' blocks. Uses double quotes ("=https") to properly nest
inside the single-quoted su -c argument while maintaining protocol
restriction.
Agent: security-auditor
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
---------
Co-authored-by: B <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
The DigitalOcean OAuth flow printed two near-identical fallback URL
messages: one manually before calling openBrowser(), and one from
openBrowser() itself. Remove the manual one since openBrowser()
already handles the fallback.
Fixes#2140
Co-authored-by: B <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
After showing post-session messages, the local process now exits cleanly
instead of requiring an extra Ctrl+C. The root cause was that after main()
resolved, lingering event loop handles (from @clack/prompts stdin listeners,
fetch connections, etc.) prevented Node/Bun from exiting naturally.
The fix adds process.exit(0) on successful main() completion, which covers
all session paths (bash script execution via execScript, SSH reconnection
via cmdConnect, and agent re-entry via cmdEnterAgent).
Fixes#2145
Agent: code-health
Co-authored-by: B <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
* test: add coverage for 6 untested pure utility functions in shared/ui.ts
Adds tests for validateServerName, validateRegionName, validateModelId,
toKebabCase, sanitizeTermValue (security-critical), and jsonEscape.
These exported functions previously had zero test coverage.
Agent: test-engineer
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* style: apply biome formatting to ui-utils test file
Address formatting review feedback: reformats destructuring import
to match project style.
Agent: pr-maintainer
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---------
Co-authored-by: B <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Remove `cleanup_stale_apps()` in `sh/e2e/lib/cleanup.sh` which was dead
code — defined but never called. The E2E orchestrator (`e2e.sh`) invokes
`cloud_cleanup_stale` directly on the active cloud driver; the wrapper
function and its file served no purpose.
Also remove the corresponding `source` call in `e2e.sh`.
Co-authored-by: spawn-qa-bot <qa@openrouter.ai>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-authored-by: L <6723574+louisgv@users.noreply.github.com>
- Remove cmdlast "should not call cmdRun when no history exists" test which
admitted in its own comment that it could not verify its stated intent and
simply duplicated the assertion from the previous test in the same describe block.
- Fix always-pass risk in manifest-type-contracts: "Interactive prompts
structure" and "Config files structure" tests iterated over optional agent
fields with a bare continue when the field was absent, meaning both tests
would vacuously pass if no agents had those fields. Added guard assertions
(expect(length).toBeGreaterThan(0)) matching the pattern used by sibling
tests in the same file.
Co-authored-by: spawn-qa-bot <qa@openrouter.ai>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
The "should use fresh disk cache without calling fetch" test only checked
toHaveProperty("agents"), which would pass even if fetch was called again.
Renamed to reflect actual behavior (in-memory cache path) and added
assertions: expect(m2).toBe(m1) and fetch call count unchanged.
Co-authored-by: spawn-qa-bot <qa@openrouter.ai>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix: address 4 reliability issues across codebase
1. sprite.ts: add --force to destroy command (stdin is "ignore" so
interactive prompts would hang until 60s timeout)
2. verify.sh: replace /dev/tcp port checks with ss -tln primary
(Debian/Ubuntu bash compiled without /dev/tcp support)
3. verify.sh: make _openclaw_restart_gateway a hard failure instead
of log_warn (matching _openclaw_ensure_gateway behavior)
4. agent-setup.ts: add ss -tln port check + "already running" early
exit + increase timeout from 120s to 300s (gateway takes ~3min
to initialize on AWS medium instances)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix: biome format - use consistent double quotes in portCheck
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: L <6723574+louisgv@users.noreply.github.com>
- sprite/sprite.ts: Replace duplicate saveVmConnection implementation
with a call to the shared saveVmConnection from history.ts. The local
version duplicated the mkdir + writeFileSync logic already provided by
the shared function, just with Sprite-specific hardcoded values.
Remove now-unused writeFileSync, mkdirSync, and getSpawnDir imports.
- Bump CLI version 0.12.5 → 0.12.6 (patch)
Co-authored-by: spawn-qa-bot <qa@openrouter.ai>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
The "should use disk cache when fresh" test in manifest.test.ts set up
a mock fetch with a comment saying it "should not be called" but never
asserted expect(global.fetch).not.toHaveBeenCalled(). The test passed
whether or not the cache was actually used, providing no signal.
-- qa/dedup-scanner
Co-authored-by: spawn-qa-bot <qa@openrouter.ai>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(e2e): improve openclaw reliability on AWS and other clouds
Three changes to make openclaw e2e tests more robust:
1. Increase PROVISION_TIMEOUT from 480s to 720s — AWS cloud-init
for "full" tier (Node.js + Bun + build-essential) can exceed 480s,
causing the CLI to be killed before .spawnrc is written.
2. Add .spawnrc manual fallback in provision.sh — if the CLI is killed
before writing .spawnrc, construct it via SSH using OPENROUTER_API_KEY
with agent-specific env vars (openclaw, zeroclaw).
3. Add retry logic to openclaw gateway input test — the gateway can
crash with 1006 websocket closure on resource-constrained instances.
Now retries once after killing and restarting the gateway process.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(security): fix command injection in e2e provision scripts
- Use printf %q and temp file for api_key handling in provision.sh to
prevent shell metachar injection (single quotes, backticks, $)
- Double-quote env_b64 interpolation in cloud_exec call to prevent
word splitting
- Replace echo with printf in bashrc append to avoid portability issues
- Replace overbroad pkill -f 'openclaw gateway' in verify.sh with
PID-targeted kill via lsof/fuser on port 18789
Agent: pr-maintainer
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: B <6723574+louisgv@users.noreply.github.com>
* test: Remove duplicate and theatrical tests
Remove 18 duplicate tests from run-path-credential-display.test.ts
that repeated coverage already provided by dedicated test files:
- "entity validation for run path" (7 tests) duplicated check-entity.test.ts
- "key resolution for run path" (6 tests) duplicated fuzzy-key-matching.test.ts
- "run-path validation sequence integration" (5 tests) duplicated
check-entity.test.ts, fuzzy-key-matching.test.ts, and script-failure-guidance.test.ts
Replace the three duplicate describe blocks with a focused 2-test
describe("isRetryableExitCode") block that covers the only unique
assertions in that section. Also remove unused spyOn import and
unused mockExit variable.
Bump version 0.12.4 → 0.12.5.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(fmt): collapse import to single line for biome format compliance
Agent: team-lead
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
---------
Co-authored-by: spawn-qa-bot <qa@openrouter.ai>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-authored-by: B <6723574+louisgv@users.noreply.github.com>
The SC2154 (referenced but not assigned) comment was leftover from a
prior version of the script. No such external variable is referenced in
the current implementation, making the suppression comment stale.
Co-authored-by: spawn-qa-bot <qa@openrouter.ai>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Add logStepInline/logStepDone helpers to ui.ts and convert all 9
polling loops (DO droplet, DO cloud-init, AWS instance, AWS cloud-init,
Hetzner cloud-init, Daytona SSH, Sprite connectivity, GCP startup,
shared SSH port) from multi-line spam to a single line that updates
in place.
Signed-off-by: Ahmed Abushagur <ahmed@abushagur.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: L <6723574+louisgv@users.noreply.github.com>
14 agent shim scripts in sh/aws/ and sh/hetzner/ were missing error
handlers on the curl command that downloads the JS bundle from GitHub
releases. If the download failed (network issue, 404, etc.), the script
would silently proceed to exec an empty/corrupt file via bun, producing
a confusing error instead of a clear "Failed to download" message.
All other clouds (GCP, Daytona, DigitalOcean, Sprite) already had this
error handling pattern. This brings AWS and Hetzner into consistency.
Agent: code-health
Co-authored-by: B <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
Adds validateRegionName() check immediately wherever awsRegion is
assigned from environment variables, rather than waiting until
createInstance(). Prevents malicious region values from being used
in SigV4 signing and shell commands.
Fixes#2113
Agent: code-health
Co-authored-by: B <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
Removes eval-based function creation pattern in e2e/lib/common.sh.
Uses variable indirection (ACTIVE_CLOUD global + wrapper functions)
instead of eval to reduce attack surface.
Fixes#2118
Agent: code-health
Co-authored-by: B <6723574+louisgv@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
