Skip to content

Commit 6cd666b

Browse files
committed
chore: merge main into inline remote header fix
2 parents ff6ff20 + 5ef5685 commit 6cd666b

3,277 files changed

Lines changed: 225661 additions & 75714 deletions

File tree

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

‎.agents/skills/test-t3-app/SKILL.md‎

Lines changed: 13 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -21,6 +21,19 @@ Test with meaningful project and thread data. Read
2121
[references/sqlite-fixtures.md](references/sqlite-fixtures.md) only when
2222
inspecting or seeding SQLite. Stop the test server before direct fixture writes.
2323

24+
## Drive agent turns for free
25+
26+
Do not spend a paid model to see a turn render. `vp run dev` registers the
27+
`Replay (dev)` provider, which replays recorded turns with simulated streaming
28+
through the real adapter and orchestration. Its models are scenarios:
29+
`showcase` covers thinking, a plan, tool calls, an approval, and markdown.
30+
Press `Ctrl+Alt+Shift+R` for a picker (GNOME takes that key for screen
31+
recording), call `window.t3dev.replay("showcase")`
32+
with `preview_evaluate`, or run
33+
`node apps/server/src/bin.ts thread start --base-dir .t3 --provider devReplay --model showcase <project> "go"`.
34+
When the test needs real provider behavior, use that provider's cheapest
35+
model, such as Haiku for Claude or Luna for Codex.
36+
2437
## Use the Browser panel
2538

2639
Call `preview_status`, then `preview_open` if the Browser panel is

‎.agents/skills/test-t3-mobile/SKILL.md‎

Lines changed: 13 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -31,11 +31,21 @@ when inspecting or seeding SQLite. Stop the test server before fixture writes.
3131
From the checkout being tested on the selected device host, run:
3232

3333
```bash
34-
node scripts/mobile-native-client.ts ensure <ios|android> <device-id>
34+
vp run mobile:install <ios|android> <device-id>
3535
```
3636

37-
This reuses a matching native client or builds and installs one. Authorized
38-
mobile verification includes that build step unless the user prohibits it.
37+
This installs the dev client for the checkout's native fingerprint: from the
38+
device, the local cache, or the shared cache in seconds, building once only
39+
when no machine has that fingerprint yet (5-15 minutes, logged to the file it
40+
names). The device id may be omitted when exactly one device is booted. It
41+
prints one JSON line; on failure it names the next command. Mobile verification
42+
includes that step unless the user prohibits it.
43+
44+
For a device you opened in an agent-device session elsewhere (a device pool,
45+
cloud, or remote host), run `vp run mobile:install <ios|android> --session <s>`
46+
from the directory you opened the session in. It installs the cached build into
47+
that session (`--abi arm64-v8a` for an ARM Android device; `--metro <url>` also
48+
opens it against Metro) and never builds.
3949

4050
Start `vp run dev:client` from `apps/mobile`, or reuse a healthy Metro belonging
4151
to this checkout. Open its printed development-client URL with AgentDevice

‎.coderabbit.config.ts‎

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -26,13 +26,19 @@ export default defineConfig((ctx) => ({
2626
path: "apps/web/src/**/*.{tsx,css}",
2727
instructions: "Hold changed code to the rules in docs/internals/web-ui.md.",
2828
},
29+
{
30+
path: "**/*.test.*",
31+
instructions:
32+
"Hold changed tests to the rules in docs/internals/testing.md. Request tests only as those rules allow.",
33+
},
2934
],
3035
},
3136
knowledge_base: {
3237
code_guidelines: {
3338
filePatterns: [
3439
{ files: "docs/internals/effect-services.md", applyTo: "{apps,packages,infra}/**/*.ts" },
3540
{ files: "docs/internals/web-ui.md", applyTo: "apps/web/src/**/*.{tsx,css}" },
41+
{ files: "docs/internals/testing.md", applyTo: "{apps,packages,infra}/**/*.{ts,tsx}" },
3642
],
3743
},
3844
},

‎.github/VOUCHED.td‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -59,6 +59,7 @@ github:myacoub91
5959
github:naMqe-h
6060
github:nateEc
6161
github:naveed949
62+
github:nexxeln
6263
github:nmggithub
6364
github:Noojuno
6465
github:notkainoa

‎.github/actions/setup-ci-deps/action.yml‎

Lines changed: 97 additions & 11 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,18 @@
11
name: Setup CI dependencies
22
description: Install Vite+ and the workspace dependencies from cache.
3+
inputs:
4+
task-cache-read-token:
5+
description: Read-only credential for the private maintainer task cache
6+
required: false
7+
default: ""
8+
install:
9+
description: Set to "false" to set up Vite+ only, with no pnpm store or install
10+
required: false
11+
default: "true"
12+
install-args:
13+
description: Arguments for `vp install`, such as filters. Empty installs the whole workspace.
14+
required: false
15+
default: ""
316
runs:
417
using: composite
518
steps:
@@ -14,10 +27,11 @@ runs:
1427
echo "node=$(jq -r '.engines.node' package.json)" >> "$GITHUB_OUTPUT"
1528
echo "pnpm=$(jq -r '.packageManager' package.json)" >> "$GITHUB_OUTPUT"
1629
30+
# Pinned because release-desktop.yml signs builds with this action in the job.
1731
- name: Cache Vite+
1832
id: vp_cache
1933
if: steps.vp.outputs.version != ''
20-
uses: actions/cache@v6
34+
uses: actions/cache@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6
2135
with:
2236
path: ~/.vite-plus
2337
key: vite-plus-home-v2-${{ runner.os }}-${{ runner.arch }}-${{ steps.vp.outputs.version }}-node${{ steps.vp.outputs.node }}-${{ steps.vp.outputs.pnpm }}
@@ -33,35 +47,107 @@ runs:
3347
echo "$HOME/.vite-plus/bin" >> "$GITHUB_PATH"
3448
"$HOME/.vite-plus/bin/vp" env use "${{ steps.vp.outputs.node }}"
3549
50+
# vp gives each Node.js download attempt 10 minutes by default, so one
51+
# stalled stream holds the job that long. It retries 3 times and resumes
52+
# partial bytes, so a shorter limit costs a slow download only a retry.
3653
- name: Setup Vite+
3754
if: steps.vp_cache.outputs.cache-hit != 'true'
38-
uses: voidzero-dev/setup-vp@v1
55+
uses: voidzero-dev/setup-vp@250f29ce396baf5e8f24498e17c0dfdebabc26eb # v1
56+
env:
57+
VP_DOWNLOAD_TIMEOUT: "120"
3958
with:
4059
node-version-file: package.json
4160
cache: false
4261
run-install: false
4362

44-
- name: Resolve pnpm store path
63+
# A shared key lets whichever install saves first, often a small filtered
64+
# one, become every job's store. Each set of install args gets its own.
65+
- name: Resolve pnpm store
4566
id: pnpm_store
67+
if: inputs.install == 'true'
4668
shell: bash
69+
env:
70+
INSTALL_ARGS: ${{ inputs.install-args }}
4771
run: |
48-
store_path="$(vp pm cache dir)"
49-
echo "path=$store_path" >> "$GITHUB_OUTPUT"
72+
echo "path=$(vp pm cache dir)" >> "$GITHUB_OUTPUT"
73+
metadata_dir="${pnpm_config_cache_dir:-$RUNNER_TEMP/pnpm-metadata}"
74+
echo "pnpm_config_cache_dir=$metadata_dir" >> "$GITHUB_ENV"
75+
echo "verified=$metadata_dir/lockfile-verified.jsonl" >> "$GITHUB_OUTPUT"
76+
if [[ -z "$INSTALL_ARGS" ]]; then
77+
echo "namespace=full" >> "$GITHUB_OUTPUT"
78+
else
79+
echo "namespace=filtered-$(printf '%s' "$INSTALL_ARGS" | cksum | cut -d' ' -f1)" >> "$GITHUB_OUTPUT"
80+
fi
5081
51-
# setup-vp's shared cache key ignores install filters. Keep full installs in a
52-
# separate namespace, including fallback restores.
82+
# Filtered installs prefer the current full store, which holds every package,
83+
# over an older store of their own.
5384
# The pnpm store stays a cache tarball on the runner's own disk. A store on a
5485
# Blacksmith sticky disk mounts in under a second, but pnpm then copies every
5586
# package across filesystems instead of hard-linking it: install took 28-49s
5687
# against 9s plus an 11s restore.
57-
- name: Cache full pnpm store
58-
uses: actions/cache@v6
88+
- name: Restore pnpm store
89+
id: pnpm_cache
90+
if: inputs.install == 'true'
91+
uses: actions/cache/restore@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6
5992
with:
6093
path: ${{ steps.pnpm_store.outputs.path }}
61-
key: pnpm-full-v1-${{ runner.os }}-${{ runner.arch }}-${{ hashFiles('pnpm-lock.yaml', 'pnpm-workspace.yaml', 'package.json') }}
94+
key: pnpm-${{ steps.pnpm_store.outputs.namespace }}-v1-${{ runner.os }}-${{ runner.arch }}-${{ hashFiles('pnpm-lock.yaml', 'pnpm-workspace.yaml', 'package.json') }}
6295
restore-keys: |
96+
pnpm-full-v1-${{ runner.os }}-${{ runner.arch }}-${{ hashFiles('pnpm-lock.yaml', 'pnpm-workspace.yaml', 'package.json') }}
97+
pnpm-${{ steps.pnpm_store.outputs.namespace }}-v1-${{ runner.os }}-${{ runner.arch }}-
6398
pnpm-full-v1-${{ runner.os }}-${{ runner.arch }}-
6499
100+
# pnpm accepts an earlier pass for the same lockfile and policy at any age, so
101+
# a cached record skips fetching registry metadata for every package.
102+
- name: Restore lockfile verification
103+
id: pnpm_verified
104+
if: inputs.install == 'true'
105+
uses: actions/cache/restore@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6
106+
with:
107+
path: ${{ steps.pnpm_store.outputs.verified }}
108+
key: pnpm-verified-v1-${{ hashFiles('pnpm-lock.yaml', 'pnpm-workspace.yaml', 'package.json') }}
109+
65110
- name: Install dependencies
111+
if: inputs.install == 'true'
112+
shell: bash
113+
env:
114+
INSTALL_ARGS: ${{ inputs.install-args }}
115+
# Unquoted so each filter is its own argument.
116+
run: vp install $INSTALL_ARGS
117+
118+
- name: Save lockfile verification
119+
if: inputs.install == 'true' && steps.pnpm_verified.outputs.cache-hit != 'true'
120+
uses: actions/cache/save@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6
121+
with:
122+
path: ${{ steps.pnpm_store.outputs.verified }}
123+
key: ${{ steps.pnpm_verified.outputs.cache-primary-key }}
124+
125+
# A filtered install that started from the current full store adds nothing
126+
# worth a second copy of it.
127+
- name: Save pnpm store
128+
if: >-
129+
inputs.install == 'true' &&
130+
steps.pnpm_cache.outputs.cache-hit != 'true' &&
131+
(steps.pnpm_store.outputs.namespace == 'full' ||
132+
steps.pnpm_cache.outputs.cache-matched-key != format('pnpm-full-v1-{0}-{1}-{2}', runner.os, runner.arch, hashFiles('pnpm-lock.yaml', 'pnpm-workspace.yaml', 'package.json')))
133+
uses: actions/cache/save@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6
134+
with:
135+
path: ${{ steps.pnpm_store.outputs.path }}
136+
key: ${{ steps.pnpm_cache.outputs.cache-primary-key }}
137+
138+
- name: Enable shared task cache
139+
if: >-
140+
github.repository == 'pingdotgg/t3code' &&
141+
(github.event_name != 'pull_request' || github.event.pull_request.head.repo.full_name == github.repository) &&
142+
hashFiles('.task-cache.json') != '' &&
143+
inputs.task-cache-read-token != ''
66144
shell: bash
67-
run: vp install
145+
env:
146+
T3_TASK_CACHE_READ_TOKEN: ${{ inputs.task-cache-read-token }}
147+
run: |
148+
node infra/task-cache/scripts/install-ci.ts
149+
if [[ ( "$GITHUB_EVENT_NAME" == pull_request && "$GITHUB_BASE_REF" == main ) || ( "$GITHUB_EVENT_NAME" == push && "$GITHUB_REF" == refs/heads/main ) ]]; then
150+
echo 'VP_REMOTE_CACHE=read-write' >> "$GITHUB_ENV"
151+
else
152+
echo 'VP_REMOTE_CACHE=read' >> "$GITHUB_ENV"
153+
fi

‎.github/scripts/ci-changes.cjs‎

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -32,6 +32,8 @@ const RUST = /^native\/(resource-monitor|kde-snap-shot|hyprland-snap-shot)\//;
3232

3333
// These tests read other packages without declaring dependencies.
3434
const TEST_ONLY_DEPS = {
35+
// Evals import the server's source to test its real model-facing text.
36+
"@t3tools/evals": ["t3"],
3537
"@t3tools/provider-pi": ["@t3tools/client-runtime"],
3638
"@t3tools/scripts": [
3739
"@t3tools/desktop",

‎.github/triage/PLAYBOOK.md‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -11,8 +11,8 @@ provided alongside this playbook. Everything machine-specific lives there, not h
1111

1212
## 1. Ask what went wrong
1313

14-
Your first message to the user: ask them to describe what went wrong, in their own
15-
words. Ask them to paste screenshots directly into this session if they have any.
14+
Unless the user already described it, your first message asks what went wrong, in
15+
their own words. Ask them to paste screenshots directly into this session if they have any.
1616
Ask follow-up questions when the description is vague. Good repro steps are the most
1717
valuable thing you can extract from this conversation.
1818

Lines changed: 55 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,55 @@
1+
name: Agent Context Budget
2+
3+
# Fails when a PR raises a value in agentContextBudget.json above main's. AgentContextBudget.test.ts
4+
# keeps that file equal to the measured context, so the context cannot grow by editing the file.
5+
# To accept growth on purpose, a maintainer adds the agent-context-increase label.
6+
on:
7+
pull_request:
8+
types: [opened, synchronize, reopened, labeled, unlabeled]
9+
10+
permissions:
11+
contents: read
12+
13+
concurrency:
14+
group: agent-context-budget-${{ github.event.pull_request.number }}
15+
cancel-in-progress: true
16+
17+
jobs:
18+
budget:
19+
name: Agent context budget
20+
runs-on: ubuntu-24.04
21+
timeout-minutes: 5
22+
env:
23+
BUDGET: apps/server/src/provider/agentContextBudget.json
24+
ALLOW_INCREASE: ${{ contains(github.event.pull_request.labels.*.name, 'agent-context-increase') }}
25+
steps:
26+
- name: Checkout
27+
uses: actions/checkout@v6
28+
with:
29+
sparse-checkout: apps/server/src/provider/agentContextBudget.json
30+
sparse-checkout-cone-mode: false
31+
32+
- name: Compare with main
33+
run: |
34+
git fetch --quiet --depth=1 origin main
35+
if ! git cat-file -e "FETCH_HEAD:$BUDGET" 2>/dev/null; then
36+
echo "main has no agent context budget yet."
37+
exit 0
38+
fi
39+
git show "FETCH_HEAD:$BUDGET" > "$RUNNER_TEMP/main.json"
40+
grown=$(jq -r --slurpfile main "$RUNNER_TEMP/main.json" '
41+
to_entries[]
42+
| select($main[0][.key] != null and .value > $main[0][.key])
43+
| "\(.key): \($main[0][.key]) -> \(.value) (+\(.value - $main[0][.key]))"
44+
' "$BUDGET")
45+
if [ -z "$grown" ]; then
46+
echo "Agent context is not larger than main."
47+
exit 0
48+
fi
49+
printf 'Agent context grew compared to main:\n%s\n' "$grown"
50+
if [ "$ALLOW_INCREASE" = "true" ]; then
51+
echo "Accepted by the agent-context-increase label."
52+
exit 0
53+
fi
54+
echo "::error file=$BUDGET::Agent context grew compared to main. Cut the added text and rerun the evals that cover it (evals/README.md), or ask a maintainer to add the agent-context-increase label."
55+
exit 1

0 commit comments

Comments
 (0)