Fix two editor input bugs:
1. Arrow key cursor movement not visually updating (fixes#464)
The layout cache key only included {width, textVersion}. Cursor-only
moves don't change textVersion, so stale cached layout was returned
and the diff renderer skipped repaint. Added cursorLine and cursorCol
to the cache key so cursor movements invalidate the cache.
2. Shift+Enter not inserting newlines in non-kitty terminals (Zed, VS Code, etc.)
The /terminal-setup command configures terminals to send ESC+CR (\x1b\r)
for Shift+Enter. But the followUp app action (bound to alt+enter) was
intercepting \x1b\r in CustomEditor.handleInput before the editor's
newLine handler could see it — because in non-kitty terminals, \x1b\r
matches alt+enter. Now when kitty protocol is not active and \x1b\r is
received, the followUp match is skipped so it falls through to newLine.
Alt+Enter followUp still works in kitty-protocol terminals (iTerm2,
Ghostty, Kitty, WezTerm) where the key combos are distinguishable.
Co-authored-by: TÂCHES <afromanguy@me.com>
Add "success" to notify type union across ExtensionUIContext, interactive
mode, and RPC mode implementations. Fix null safety for readFileSync and
contextUsage.percent in auto.ts. Add discriminated union narrowing for
dispatch results. Add string type guards for select() return values in
commands.ts. Align ProviderErrorPauseUI notify signature. Simplify
AuthStorage return type.
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Replace existsSync collision loop with atomic O_CREAT|O_EXCL file
creation, hoist regex to module-level constant, and memoize
getPackageDir() to avoid repeated directory walks.
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Arrow keys produce `^[[D`/`^[[C` instead of moving the cursor when event
loop latency causes the StdinBuffer to split escape sequences.
Three layered fixes:
1. Increase StdinBuffer timeout from 10ms to 50ms (matches xterm default)
so split escape sequences are reassembled even under load.
2. Clean up stale readline listeners after @clack/prompts onboarding —
readline.emitKeypressEvents() leaves a permanent data listener that
is unnecessary for the TUI.
3. Guard in editor against CSI remnants: if a split still occurs, reject
text matching navigation escape patterns ([A-F, [H, [Z, [n~) instead
of inserting them as characters.
Closes#493
When auto-mode creates a worktree and chdir's into it, the Node process
cwd changes but AgentSession._cwd stays frozen at the original path.
Every newSession() builds a system prompt telling the LLM "Current
working directory: /original/path", so the LLM cd's back there and
writes files to the wrong location.
Update _cwd = process.cwd() at the start of newSession() so the system
prompt reflects the actual working directory after chdir.
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* docs: add startup performance analysis and optimization plan
Profiled GSD CLI startup finding 2.2s for --version and ~3.8s for
interactive mode. Identified 5 root causes with measured timings and
created a phased optimization plan targeting <0.2s for --version
and ~0.8s for interactive startup.
* perf: speed up GSD startup with lazy loading and fast paths
- Fast-path --version/-v and --help/-h in loader.ts before importing
any heavy dependencies (2.2s → 0.15s, 14x faster)
- Lazy-load undici (~200ms) only when HTTP_PROXY env vars are set
- Skip initResources cpSync when managed-resources.json version
matches current GSD version (~128ms saved per launch)
- Lazy-load Mistral SDK (~369ms) on first API call instead of startup
- Lazy-load Google GenAI SDK (~186ms) on first API call instead of
startup
- Parallelize extension loading with Promise.all() instead of
sequential for-loop
---------
Co-authored-by: TÂCHES <afromanguy@me.com>
First test coverage for pi-tui components (8 tests):
- Loader: start() idempotency, dispose() cleanup, stop() safety
- CancellableLoader: abort on dispose, callback cleanup, signal state
- Input: paste buffer reset on focus loss, focused getter/setter
- Loader: clear existing interval in start() to prevent orphaned timers
on double-call; add dispose() to stop and null TUI ref
- CancellableLoader: abort the AbortController and clear onAbort in
dispose() so external signal holders release the controller
- Editor: add public dispose() that clears the autocomplete debounce
timer to prevent post-removal callbacks
- Input: convert focused to getter/setter that resets isInPaste and
pasteBuffer on focus loss, preventing paste state corruption
- TUI.stop(): iterate overlayStack calling dispose() on each component
before clearing, stopping overlay timers (e.g. dashboard refresh)
The Input component had no placeholder text support — when empty, it
showed only "> " with a blinking cursor and no hint of expected input.
The ExtensionInputComponent received a placeholder parameter but
discarded it (_placeholder with underscore = intentionally unused).
Fix: Input now has a public placeholder property. When value is empty,
renders the placeholder in dim text. ExtensionInputComponent passes
the placeholder through to Input.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Split skill diagnostics into [Skill conflicts] (actual collisions) and
[Skill issues] (validation warnings like missing description) so users
aren't misled by the label. Add wizard hint to /gsd prefs output.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The synchronous fuzzyFind() native call blocks the event loop during
@ file autocomplete. On large codebases (e.g. Java projects with deep
directory trees), each call can take seconds. Since updateAutocomplete()
was called on every keystroke while autocomplete was active, rapid typing
would cascade into dozens of blocking searches — freezing the TUI for
minutes. This made it appear that arrow keys caused the freeze, when
the actual cause was accumulated backlog from processing buffered input.
Debounce all @ file reference autocomplete paths (character input,
backspace, forward delete, and re-trigger after cancellation) with a
150ms delay so only the final keystroke triggers the expensive search.
Slash command autocomplete remains synchronous since it's cheap.
Co-authored-by: TÂCHES <afromanguy@me.com>
GitHub Copilot users with Claude models got 400 errors because the native
Anthropic web_search_20250305 tool was injected into requests to Copilot's
API proxy, which doesn't support it. The root cause was that model_select
never fires before the first API request on new sessions, so the fallback
heuristic (model name starts with "claude-") couldn't distinguish direct
Anthropic from proxied providers.
Fix: pass the resolved Model object through to the before_provider_request
event so extensions can check model.provider directly instead of relying on
model name heuristics.
Moves extension tool_call/tool_result interception from wrapToolsWithExtensions
(which fires inside the agent loop, bypassing event settlement) to
beforeToolCall/afterToolCall hooks that await _agentEventQueue. This ensures
extensions always see settled state — including the appended assistant message —
even when tools execute in parallel.
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Add Ollama Cloud (ollama.com) as a built-in provider with both model
hosting and web search/fetch capabilities.
Model provider:
- 13 curated models via OpenAI-compatible API (Llama 3.1, Qwen 3,
DeepSeek R1, Gemma 3, Mistral, Phi-4, GPT-OSS)
- Auth via OLLAMA_API_KEY environment variable
- Registered in onboarding, env hydration, and model resolver
Web tool provider:
- Search via POST ollama.com/api/web_search
- Page fetch via POST ollama.com/api/web_fetch (fallback after Jina)
- Added as third search provider option alongside Tavily and Brave
- /search-provider command updated with ollama option
Closes#430
When all credentials for a provider are exhausted, the system now
automatically falls back to the next available provider in a
user-configured fallback chain. Higher-priority providers are
restored automatically when their backoff expires.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Adds a "Custom (OpenAI-compatible)" provider option to the API key
flow in the onboarding wizard. When selected, prompts for base URL,
API key, and model ID, then writes the config to models.json.
* fix(auto): prevent hang when dispatch chain breaks after slice tasks complete (#381)
After the last task in a slice completes, dispatchNextUnit() can throw
(e.g. template mismatch, branch error, or any unprotected operation).
The error propagates to the pi event emitter which silently swallows
async rejections, leaving auto-mode active but permanently stalled —
no dispatch, no stop, no recovery.
Three defensive layers added:
1. Try-catch around dispatchNextUnit in handleAgentEnd — catches errors,
shows them to the user, and schedules a retry via the gap watchdog.
2. Dispatch gap watchdog (30s timer) — fires when auto-mode is active
but no unit was dispatched after a unit completion. Re-derives state
and retries. If retry fails, stops auto-mode with diagnostics.
3. Error boundary in the agent_end event handler — last-resort catch
that pauses auto-mode if handleAgentEnd itself throws.
Closes#381
* fix: improve Cloud Code Assist 404 error with actionable model guidance (#368)
When a model like gemini-2.0-flash isn't available via Cloud Code Assist,
the 404 error now names the model and suggests using the google provider
with GOOGLE_API_KEY or switching to a supported model.
* fix: remove @gsd/* cross-deps that break npm install (#hotfix)
Workspace packages declared @gsd/* as dependencies in their own
package.json files. npm's bundleDependencies bundles packages into
node_modules/ but still tries to resolve sub-dependencies from the
registry — causing 404s for the unpublished @gsd/* scope.
- Remove @gsd/* from all dependencies (root and workspace packages)
- Add validate-pack.sh: tests tarball installability before publish
- Wire validate-pack into CI (every PR) and publish pipeline
- Bump to v2.10.10
- Update changelog
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: drop bundleDependencies, use postinstall symlinks instead
bundleDependencies with workspace packages causes npm to resolve
@gsd/* from the registry during install — 404 since they're not
published. Replace with a postinstall script that creates
node_modules/@gsd/* symlinks pointing to packages/*.
- Remove @gsd/* from dependencies and bundleDependencies
- Add link-workspace-packages.cjs (CJS, runs before ESM postinstall)
- Update validate-pack to verify symlinks after install
- Include link script in files array
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: robust validate-pack + fallback workspace linking
- Keep @gsd/* in bundleDependencies (for npm pack bundling)
- Remove @gsd/* from root dependencies (prevents 404 registry lookups)
- Add link-workspace-packages.cjs fallback for when bundled symlinks
aren't created
- Simplified validate-pack with better error diagnostics
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: remove bundleDependencies — use postinstall symlinks only
npm 10.x fetches packument metadata for ALL deps including bundled ones.
@gsd/* packages don't exist on npm → 404 → hard install failure.
bundleDependencies is fundamentally broken for unpublished workspace
packages. Replace with:
- packages/ shipped via files array (already was)
- link-workspace-packages.cjs creates node_modules/@gsd/* symlinks in
postinstall, pointing to packages/*
- No @gsd/* in dependencies or bundleDependencies at all
Tarball drops from 40M to 3M (no bundled node_modules).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: add .npmignore to prevent .gitignore from excluding dist/
.gitignore contains /dist/ and packages/*/dist/ which are needed in
the published tarball. Without .npmignore, npm pack respects .gitignore
and excludes them — even though "files" in package.json should override.
An empty .npmignore causes npm to ignore .gitignore entirely, letting
the "files" field control what's packed.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: avoid SIGPIPE in validate-pack on Linux
tar | grep -q causes SIGPIPE (exit 141) on Linux when grep closes the
pipe early. Write tar listing to a temp file and grep that instead.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: normalize .ts import extensions to .js for Node 22.22+ compatibility
Node 22.22.0's --experimental-strip-types handles .ts import specifiers
differently, causing ERR_INVALID_TYPESCRIPT_SYNTAX in CI. The project
convention uses .js specifiers with a custom resolve hook that rewrites
them to .ts at test time.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix: replace better-sqlite3 with sql.js to eliminate native compilation failures
better-sqlite3 requires prebuilt binaries or node-gyp compilation, which
fails on newer Node versions (e.g. 25.x) that lack prebuilds. sql.js uses
WASM-compiled SQLite with zero native dependencies, making installation
reliable across all platforms and Node versions.
Closes#355
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Initial plan
* Add adaptive reconciliation, concrete remediation steps, and xxHash32 JS fallback
- auto.ts: final reconciliation pass before halting on loop detection
(runs skipExecuteTask to write blocker artifacts and advance pipeline)
- auto.ts: adaptive retry on prevCount>=2: write stub summary when still
missing after two agent sessions, so the agent has a recovery context
- auto.ts: buildLoopRemediationSteps() helper with concrete manual steps
when automatic reconciliation cannot resolve the unit
- xxhash/index.ts: pure-JS xxHash32 fallback when native addon does not
export xxHash32 (prevents 'native.xxHash32 is not a function' crash)
- idle-recovery.test.ts: tests for buildLoopRemediationSteps and loop-recovery
path in skipExecuteTask
Co-authored-by: glittercowboy <186001655+glittercowboy@users.noreply.github.com>
* Address code review feedback
- xxhash/index.ts: extract accumulate() helper to remove duplicate lane
processing lines in the 16-byte loop
- auto.ts: replace magic number 2 with named constant STUB_RECOVERY_THRESHOLD
- idle-recovery.test.ts: add comment explaining the 3 (== MAX_UNIT_DISPATCHES)
used in the loop-recovery test
Co-authored-by: glittercowboy <186001655+glittercowboy@users.noreply.github.com>
* Address PR review: verify post-state, fix stub wording, fix checkbox syntax, add fallback test
- auto.ts (MAX branch): check verifyExpectedArtifact() after skipExecuteTask()
before advancing; annotate both conditions explaining what each validates
- auto.ts (self-repair branch): guard advance with verifyExpectedArtifact() so
regex mismatches in the slice plan don't silently re-dispatch
- auto.ts (stub-recovery): write stub summary directly with "PARTIAL RECOVERY"
wording instead of using skipExecuteTask (which would claim "exhausted N attempts")
- auto.ts (remediation): fix checkbox example to "- [x] **${tid}:" (no trailing **)
- xxhash/index.ts: export xxHash32Fallback so CI can test JS path independently
- native/src/index.ts: re-export xxHash32Fallback
- xxhash.test.mjs: add "xxHash32Fallback (pure-JS path)" describe block that
validates the JS implementation against the reference on all test cases
Co-authored-by: glittercowboy <186001655+glittercowboy@users.noreply.github.com>
* chore: retrigger CI
---------
Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: glittercowboy <186001655+glittercowboy@users.noreply.github.com>
Co-authored-by: CI Trigger <noreply@github.com>
Model variants like `claude-opus-4-6[1m]` use bracket suffixes to
differentiate context window configurations internally, but the
Anthropic API only accepts base model IDs (e.g. `claude-opus-4-6`).
Sending the full variant ID via OAuth (Claude Max/Pro) causes a 404:
{"type":"not_found_error","message":"model: claude-opus-4-6[1m]"}
Strip any `[...]` suffix from model.id for OAuth requests only.
API key auth is left unchanged since the behavior there is unverified.
On macOS the Alt key is the Option key (⌥), but all UI hints showed
"Alt+". Added formatKeyForDisplay() utility that converts alt+ to ⌥
on darwin, applied it in formatKeys() for dynamic keybinding hints,
and updated hardcoded strings in tree-selector, models-selector,
settings-selector, auto-mode dashboard, and extension shortcut display.
Hashline prefixes (e.g. "1#BQ:") were leaking into the TUI display
for file reads, showing as weird characters to users. Strip them
before rendering since they're only meant for model consumption.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Anthropic's 429 responses include retry-after and x-ratelimit-reset-*
headers that tell us exactly when to retry. Previously we ignored these
and used exponential backoff (2s, 4s, 8s), which is both wrong and
misleading in the UI countdown.
- Add retryAfterMs to AssistantMessage as the structured carrier
- Extract retry-after / x-ratelimit-reset-requests / x-ratelimit-reset-tokens
from Anthropic SDK APIError.headers in the provider catch block
- Session uses retryAfterMs when present (capped by maxDelayMs=60s),
falls back to exponential backoff for errors with no timing hint
The UI countdown now shows the actual Anthropic reset time. No UI changes needed.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
