v0.6.3

What's Changed

• ci(loupe): run loupe from main by @atbe in #144
• fix(tui): stop viewport oscillation during markdown streaming by @anish-inf in #146

Full Changelog: v0.6.2...v0.6.3

v0.6.2

What's Changed

• fix(tui): keep inline view bottom-anchored by @anish-inf in #142
• fix(tui): bottom-anchor opencode view so the final markdown flush doesn't jump by @anish-inf in #143

Full Changelog: v0.6.0...v0.6.2

whipcode v0.0.10

Whipcode branch build from whip-rlm.

Source commit: 2cea57c23b138284974226e9175d8bf1d8c9ba97

Install:

curl -fsSL https://raw.githubusercontent.com/context-labs/whip/whip-rlm/install-whipcode.sh | sh

Whip Desktop 0.2.0 Beta 2

Whip Beta for Apple Silicon Macs running macOS 14 or newer. This release includes the desktop app and its matching whipcode backend.

Install

1. Download Whip-Beta-0.2.0-beta.2-arm64.dmg from the assets below.
2. Open the DMG and drag Whip Beta.app into Applications.
3. Open Whip Beta and connect This Mac. On a new installation, choose Install whipcode and save it to the suggested location.
4. Configure your model provider in Settings and start a session.

The suggested CLI location is ~/.local/bin/whipcode. Add ~/.local/bin to your shell's PATH if you also want to run whipcode from Terminal.

Updates

The app checks the beta update channel and downloads updates in the background. Choose Restart to update to apply an update to the app and its managed local backend together. Restarting the backend interrupts active work; sessions and configuration remain on disk.

An existing independently installed binary is treated as externally managed. Whip does not silently replace it. SSH and URL hosts also require an explicit backend upgrade.

Downloads and verification

• DMG: normal Mac installation.
• ZIP: signed application archive used for automatic updates.
• whipcode-linux-x64: matching standalone backend for Linux x64 execution hosts.
• SHA256SUMS, artifact-manifest.json, package/startup evidence, dependency inventory, and third-party notices accompany the release.

Artifacts are signed with Developer ID and notarized by Apple. GitHub attestations bind the files to the release source and workflow.

This is a beta release. Intel Macs and Windows desktop builds are not included.

Release acceptance

Release build and provenance and actual signed app/backend update acceptance.

The exact candidate passed 61 launch measurements on macOS 14 and macOS 15 CI runners. An actual Squirrel update from beta.0 to beta.2 verified that downloading and cancellation preserve running work; one confirmation updates the app and managed backend; sessions/configuration survive; interruption is recorded; and Unix/WebSocket clients reconnect.

The supplemental acceptance-macos14.json contains the downloaded archive identity, launch measurements, and updater results. Its SHA-256 is 369b83a632b44df611358af1046e5303fc0efa16df6c93710e531bbfe69d45b3. Build artifacts are covered by the attested candidate manifest; this supplemental report is linked to the acceptance run above.

macOS 14 coverage used an Apple Silicon CI VM. Physical-device permission and compatibility QA remains on the stable-release checklist.

v0.6.1

What's Changed

• chore(deps): bump the gomod-minor-patch group with 2 updates by @dependabot[bot] in #126
• chore(deps): bump Infisical/secrets-action from 1.0.15 to 1.0.17 in the actions group by @dependabot[bot] in #127
• chore(deps): bump actions/checkout from 5 to 7 by @dependabot[bot] in #128
• chore(loupe): halve maxTurns, drop reasoning one tier (cost) by @atbe in #129
• Boot new installs in opencode UI mode by @anish-inf in #133
• feat: whip -c/--continue, -r/--resume shorthand, and --browse session picker by @anish-inf in #134
• fix(tui): shorten the cwd so the spend survives status-line truncation by @anish-inf in #135
• Add dynamic workflows: a workflow tool for deterministic multi-agent … by @anish-inf in #132
• Fix/ascii says whip by @anish-inf in #136
• chore(loupe): switch main review model to glm-5.3 (cost) by @atbe in #137
• Add Codex (ChatGPT subscription) provider by @atbe in #130

Full Changelog: v0.5.14...v0.6.1

v0.6.0 — Codex subscription

whip 0.6.0 adds a ChatGPT Codex subscription provider: use the models your ChatGPT plan already includes, with no API key. It also ships one shared retry policy for every provider, a friendlier model picker, session resume shorthands, and an experimental multi-agent workflow tool.

Install or upgrade

# fresh install (macOS / Linux)
curl -fsSL https://raw.githubusercontent.com/context-labs/whip/main/install.sh | sh

# already installed
whip update

# from source
go install github.com/context-labs/whip/cmd/whip@v0.6.0

Binaries for darwin/amd64, darwin/arm64, linux/amd64 and linux/arm64 are attached below with SHA256SUMS.

Use your Codex subscription

whip auth codex        # opens a device-code sign-in: visit the URL, enter the code
whip                   # /model now lists every model your plan includes

• Inside a session, /auth codex does the same. Neither flow changes your default model; pick one with /model or /model <id> codex-subscription.
• /usage shows your subscription's rate-limit windows (used %, reset countdown) without leaving whip.
• whip auth codex logout removes the provider again. whip auth openrouter logout and whip auth --help are new too.
• Login is stored in ~/.codex/auth.json, compatible with the Codex CLI; tokens refresh locally and are never sent to the conversation or to any host other than chatgpt.com/backend-api.
• Subagents on the subscription default to gpt-5.6-luna; pin any model via ctrl+p › Subagent model or taskModel in config.

Migrating a pre-release config: a provider named codex is renamed to codex-subscription on first load, along with its routes, pins and cached catalog. Nothing to do by hand.

Highlights

Codex subscription provider (#130, supersedes #14)

• New openai-codex-responses API over the Responses SSE endpoint: tool calls, reasoning replay, usage, prompt caching, streaming tool-call rows.
• Account-scoped /codex/models catalog drives /model; plan and rollout availability come from the backend, not a hard-coded list.
• Exhausted windows surface immediately as "Codex subscription limit reached … resets in …" rather than a retry loop. A 401 on a rotated token triggers one forced refresh and a resend.

One retry policy for every provider

• Transient failures (transport errors, 429, 5xx, mid-stream server_error) back off exponentially, honour Retry-After, report each retry in the UI, and never replay output already shown — including pending tool-call rows.
• A Retry-After beyond 60 seconds is treated as a quota, not congestion, and surfaces instead of stalling the turn.

Model pickers

• /model and the ctrl+p Model, Subagent model and Compaction model panels show model @ provider @ endpoint rows grouped by provider, window to the terminal height, and keep the cursor on screen while scrolling.
• Subagent and compaction models can be any configured or catalog model.

Sessions and startup

• whip -c / --continue resumes the latest session, whip -r <id> / --resume a specific one, --browse opens the session picker.
• New installs boot in the opencode UI mode.

Dynamic workflows (experimental)

• A workflow tool for deterministic multi-agent orchestration, gated behind "experimental": ["workflows"] in config.

Fixes

• Catalog-only models on a non-default provider resolve again; defaultProvider now only breaks ties between catalogs advertising the same id.
• Removing your last provider persists cleanly and clears dangling model pins.
• Tool-call bookkeeping (duration, exit code) is no longer erased from the live conversation when a request is built.
• A stream that dies mid tool-call drops that call instead of persisting malformed arguments.
• Status line keeps the spend visible on narrow terminals; several lint and test-reliability fixes.

For embedders

llm.Client is now an interface implemented by OpenAI and Codex, with Clone, SetCacheKey, Endpoint, and an optional SetOnRetry. agent.New takes the interface.

Full changelog: v0.5.14...v0.6.0

whipcode v0.0.6

Whipcode branch build from whip-rlm.

Source commit: a7e1a2e6156dd57d731a4ded93f206d823765756

Install:

curl -fsSL https://raw.githubusercontent.com/context-labs/whip/whip-rlm/install-whipcode.sh | sh

whipcode v0.0.5

Whipcode branch build from whip-rlm.

Source commit: 385b27571c41a632e5569d13e86018aa98359a03

Install:

curl -fsSL https://raw.githubusercontent.com/context-labs/whip/whip-rlm/install-whipcode.sh | sh

whipcode v0.0.4

Whipcode branch build from whip-rlm.

Source commit: 5164e0415519336e310dea50d637c3fcba69cbe9

Install:

curl -fsSL https://raw.githubusercontent.com/context-labs/whip/whip-rlm/install-whipcode.sh | sh

v0.5.14

What's Changed

• fix(tui): small-terminal scrolling — cap the live area, don't yank scroll on stream by @francescov1 in #124
• fix(mcp): codex tool-approval tables + bearer_token_env_var; add whip skills import by @francescov1 in #121
• feat(tui): render pasted images as compact [Image N] chips by @francescov1 in #123
• fix(tools): replace binary tool output with a compact [binary] placeholder by @francescov1 in #122

Full Changelog: v0.5.13...v0.5.14