mirror of
https://github.com/warpdotdev/warp.git
synced 2026-05-22 23:32:19 +08:00
master
6 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Zach Lloyd
|
424e3d827f |
Allow common skills install failures to continue scripts (#11438)
## Description - Make common-skills install failures non-fatal in `script/run` and `script/bootstrap`. - Print a clear stderr line when installation cannot complete, then continue the normal run/bootstrap flow. - Keep strict resolver/install behavior unchanged outside the entrypoint failure handling. Oz conversation: https://staging.warp.dev/conversation/2fff0589-38e0-48ec-869e-707afcc1b10b ## Linked Issue - [ ] The linked issue is labeled `ready-to-spec` or `ready-to-implement`. - [ ] Where appropriate, screenshots or a short video of the implementation are included below (especially for user-visible or UI changes). ## Testing - [x] `bash -n script/run script/bootstrap` - [x] Isolated temp-directory harness that forces `script/resolve_common_skills` to fail and confirms both entrypoints continue. - [ ] I have manually tested my changes locally with `./script/run` ## Agent Mode - [x] Warp Agent Mode - This PR was created via Warp's AI Agent Mode CHANGELOG-NONE Co-Authored-By: Oz <oz-agent@warp.dev> Co-authored-by: Oz <oz-agent@warp.dev> |
||
|
Zach Lloyd
|
a529cdbf03 |
Move common skills management to common-skills scripts (#10617)
## Description Move common-skills install/remove/verify responsibilities out of the Warp checkout and into `warpdotdev/common-skills/scripts`, then have Warp delegate to those scripts. This keeps the common skill management logic with the common skills repo while preserving the existing Warp developer flows: - `script/bootstrap` and `script/run` resolve the common-skills scripts from `WARP_COMMON_SKILLS_SCRIPTS_DIR`, a sibling `../common-skills` checkout, or a common-skills worktree. - Both entrypoints pass `--repo-root` so the external scripts operate on the current Warp checkout. - Bootstrap and run now verify installed common skills against `skills-lock.json` after install/update. - Documentation and Oz environment guidance now point at the external common-skills scripts. This PR depends on the companion `warpdotdev/common-skills` branch `zach/move-common-skill-scripts`, which adds the scripts used here. script/bootstrap <img width="839" height="182" alt="Screenshot 2026-05-10 at 3 32 42 PM" src="https://github.com/user-attachments/assets/347e469f-ac4c-4e18-980f-87d6a9a1a469" /> <img width="1044" height="102" alt="Screenshot 2026-05-10 at 5 37 22 PM" src="https://github.com/user-attachments/assets/09f6d4ff-43a5-40c8-88c7-8df82c73b74c" /> ## Linked Issue N/A - [ ] The linked issue is labeled `ready-to-spec` or `ready-to-implement`. - [ ] Where appropriate, screenshots or a short video of the implementation are included below (especially for user-visible or UI changes). ## Testing - `bash -n script/run script/bootstrap` - `bash -n /Users/zach/.warp-dev/worktrees/common-skills/move-common-skill-scripts/scripts/install_common_skills /Users/zach/.warp-dev/worktrees/common-skills/move-common-skill-scripts/scripts/remove_common_skills /Users/zach/.warp-dev/worktrees/common-skills/move-common-skill-scripts/scripts/verify_common_skills` - `pwsh -NoProfile -Command '$null = [scriptblock]::Create((Get-Content -Raw "script/windows/bootstrap.ps1"))'` - `git diff --check` - Smoke-tested bootstrap/run common-skills paths using temporary copies that no-op platform setup/app launch. - [ ] I have manually tested my changes locally with `./script/run` ### Screenshots / Videos N/A — script and documentation changes only. ## Agent Mode - [x] Warp Agent Mode - This PR was created via Warp's AI Agent Mode Conversation: https://staging.warp.dev/conversation/d6752e7c-9dfc-4929-a524-002bd7d2ecaa Co-Authored-By: Oz <oz-agent@warp.dev> --------- Co-authored-by: Oz <oz-agent@warp.dev> |
||
|
Evelyn Xu
|
fd2a608ace |
[Quality-547] add bert_tiny_v2 model and update input classifier (#10756)
## Description
<!-- Please remember to add your design buddy onto the PR for review, if
it contains any UI changes! -->
Add [retrained nld
classifier](https://github.com/warpdotdev/ml_models/pull/2) to dogfood;
Meanwhile changing model file storage from raw blobs to lfs for
optimizing git efficiency
It could resolve the misfires reported in
[quality-547](https://linear.app/warpdotdev/issue/QUALITY-547/laundry-list-of-misfires);
For quality 543 on file paths misfires, some are still pending to
resolve by improving heuristics.
## About switching to git lfs to store model binary files
| Scenario | Potential Impact | Failure Risk | Mitigation |
|---|---|---|---|
| Fresh clone with `git-lfs` installed | `git clone` (and subsequent
`git pull`) automatically invokes the smudge filter, which fetches the
real ONNX blobs from LFS in place of the pointer files. `rust-embed`
embeds the actual model bytes at build time. | None — classifier loads
normally. | No mitigation needed; this is the happy path. |
| Fresh clone without `git-lfs` installed | ONNX paths land as ~130-byte
pointer stubs. `rust-embed` embeds the pointer file at build time;
runtime tries to load it as ONNX. | `candle::InferenceRunner::new` fails
and the classifier silently falls back to `HeuristicClassifier`. App
still builds and runs — no compile-time break, but a behavioral
regression. | `script/bootstrap` now refuses to proceed if `git-lfs` is
missing and runs `git lfs install --local && git lfs pull` when present;
`WARP.md` documents it. |
| Existing clone, user pulls this branch with `git-lfs` installed but
never enabled in the repo | `git lfs install --local` hasn't been run,
so the smudge filter is inactive. | The v1 file may stay correct
(already a blob locally) but the new v2 file appears as a pointer. |
Same bootstrap step covers it; users who skip bootstrap can run `git lfs
install && git lfs pull` once. |
| Existing clone, user pulls without ever installing `git-lfs` | The v2
model lands as a pointer (silent classifier fallback). | Build succeeds,
but classifier silently degrades to heuristic fallback. | Bootstrap
requirement + `WARP.md` note + the runtime `log::warn!("Failed to load
onnx classifier (v2): ...")` makes the failure greppable in logs. |
| Existing clone with `git-lfs` installed and enabled | `git pull`
invokes the smudge filter automatically, fetching the v2 blob from LFS
to replace the pointer. `rust-embed` picks up the real bytes on the next
build. | None — classifier loads normally. | No mitigation needed; this
is the happy path for existing clones. |
| Forks / branches based on pre-LFS commits | Cherry-picks or merges
crossing the LFS boundary need `git-lfs` locally to render the binaries
correctly. | Low — no history rewrite is needed since we're only
renormalizing on this commit. | Old branches keep working with raw blobs
because their `.gitattributes` doesn't yet have the LFS rule. |
### TL;DR — Risk Mitigation Summary
**1. Fresh clones (hard enforcement at bootstrap)**
The actual gate lives in `script/bootstrap`, which now (a) refuses to
proceed if `git-lfs` is not installed, and (b) runs `git lfs install
--local && git lfs pull` automatically when it is. The `WARP.md` note
complements this by explaining the requirement and providing recovery
commands for users who bypass bootstrap.
**2. Existing clones (tiered recovery)**
- **Hard failure path:** If a build or runtime error surfaces (e.g.
`candle::InferenceRunner::new` panics or the agent reports a missing
model), the agent's remediation is to run `git lfs install --local &&
git lfs pull` once and retry.
- **Silent fallback path:** If the build succeeds but the classifier
silently regresses to `HeuristicClassifier`, we accept it as a
non-blocking degradation. The `log::warn!("Failed to load onnx
classifier (v2): ...")` line makes the condition greppable in logs, and
the next bootstrap run will self-heal.
**3. CI (requires explicit opt-in to LFS)**
Update CI `with: lfs: true` to any job that compiles the app, runs
classifier tests, or bundles release artifacts. Please see
|
||
|
Zach Lloyd
|
b84e3e98d9 |
Add opt-in common skills installer (#10121)
## Description Replaces the custom common-skills lock file with the standard project lock generated by `npx skills`, and wires local setup to restore checked-in common skills from that lock. This PR: - Removes `.agents/common-skills.lock` and the custom common-skills lock/update workflow. - Adds `skills-lock.json`, the standard `npx skills` project lock for skills sourced from `warpdotdev/common-skills`. - Keeps the copied `.agents/skills/*` files checked in so local and cloud agents can discover repo-local skills directly from the checkout. - Reworks `script/install_common_skills` to restore from `skills-lock.json` via `npx --yes skills@1.5.6 experimental_install`. - Stores the local install stamp under `.git/warp/common-skills-lock.hash` so normal restores do not dirty the worktree. - Updates `script/run` to check the lock before building and restore skills only when the lock hash changed. - Keeps `./script/bootstrap --install-common-skills` and `.\script\windows\bootstrap.ps1 -InstallCommonSkills` as explicit bootstrap paths that delegate to the same installer. - Adds `specs/common-skills-installation/TECH.md` with separate Mermaid diagrams for the local install/update flow and the Oz Claude cloud-agent environment setup flow. Pretty cool loom of what's coming on orchestration for anyone interested :) https://www.loom.com/share/87212dc5a6824573861794e32cf9902d ## Flow ### Local agent install/update flow 1. `warpdotdev/warp` checks in both `skills-lock.json` and `.agents/skills/*`. 2. `script/run` calls `script/install_common_skills --if-needed --non-interactive --quiet` before launching the local build. 3. The installer hashes `skills-lock.json` and compares it with `.git/warp/common-skills-lock.hash`. 4. If the hash matches, it skips restore and continues the build. 5. If the hash is missing or stale, it runs `npx --yes skills@1.5.6 experimental_install`, restores `.agents/skills/*` with the pinned `skills@1.5.6` CLI, writes the new local stamp, and then continues. 6. To intentionally update common skills, run `npx --yes skills@1.5.6 update -p -y` and commit the resulting `skills-lock.json` and `.agents/skills/*` changes together. ### Oz Claude cloud-agent environment setup flow 1. Oz provisions or reuses a cloud environment and clones/syncs `warpdotdev/warp`. 2. Environment setup should invoke `./script/install_common_skills --if-needed` after checkout sync and before starting the Claude agent. 3. Fresh environments restore from `skills-lock.json`; reused environments skip when the cloud-local stamp matches. 4. The Claude agent starts with repo-local checked-in skills discoverable, and runs may still pass explicit skill specs such as `warpdotdev/warp:create-pr`. The Oz environment hook itself lives outside this repo; this PR makes the repository checkout self-sufficient and idempotent when that hook invokes the installer. ## Linked Issue - [ ] The linked issue is labeled `ready-to-spec` or `ready-to-implement`. - [ ] Where appropriate, screenshots or a short video of the implementation are included below (especially for user-visible or UI changes). ## Screenshots / Videos Not applicable; script/docs/skills change. ## Testing - `bash -n script/install_common_skills script/run script/bootstrap` - PowerShell parse check for `script/windows/bootstrap.ps1` - `./script/install_common_skills --if-needed --quiet` using pinned `skills@1.5.6` - second `./script/install_common_skills --if-needed --quiet` skip-path check - `git diff --check` - Earlier branch validation: `cargo check --manifest-path Cargo.toml` - Earlier branch validation: `cargo fmt --all --manifest-path Cargo.toml` - Earlier branch validation: `cargo clippy --manifest-path Cargo.toml --workspace --all-targets --all-features --tests -- -D warnings` ## Agent Mode - [x] Warp Agent Mode - This PR was created via Warp's AI Agent Mode Co-Authored-By: Oz <oz-agent@warp.dev> --------- Co-authored-by: Oz <oz-agent@warp.dev> |
||
|
Tautik Agrahari
|
ce89a98b53 |
fix(bootstrap): warn before sudo and document install steps (#9501)
- new `script/warp_sudo`: echoes each privileged command, prompts
`[y/N]` from `/dev/tty`, shells to real `sudo` on confirm.
`WARP_BOOTSTRAP_YES=1` or non-tty stdin skips.
- `script/bootstrap`: static `--help` install list dropped. `-y`/`--yes`
exports `WARP_BOOTSTRAP_YES=1` for children.
- ten direct `sudo` call-sites in `script/{macos,linux}/bootstrap` +
`script/linux/install_{build,runtime,test}_deps` switched to
`warp_sudo`. windows unchanged.
- `README.md`: short "what `./script/bootstrap` does" subsection
pointing at the platform scripts as the source of truth.
closes #9421.
|
||
|
David Stern
|
0dbd3d567a |
Initial public release of Warp.
Repo-Sync-Origin: warpdotdev/warp-internal@12af1d983b |