chore(deps-dev): bump typescript from 5.9.3 to 7.0.2 - #15
Closed
dependabot[bot] wants to merge 73 commits into
Closed
dependabot[bot] wants to merge 73 commits into
dependabot[bot] wants to merge 73 commits into
Conversation
Ship the daemon-as-proxy, TCB-in-daemon, and CLI boot/pull work that landed after 0.1.11, and align the lockfile (was still 0.1.1). Co-authored-by: Cursor <cursoragent@cursor.com>
Capture the playbook-vs-sticky-note model so global classes/recipes and a small project policy file stay distinct. Co-authored-by: Cursor <cursoragent@cursor.com>
Add packages/core/CHANGELOG.md starting at this version (no backfill). Include CHANGELOG.md in the npm "files" array. Add a persistent .cursor/rules/npm-publish.mdc rule so every @mba-ai/* publish requires a version bump and a new changelog section. Co-authored-by: Cursor <cursoragent@cursor.com>
…p the original-project fallback Replace the hardcoded the original project default for the llama.cpp fork with a real search: MBA_LLAMA_SERVER_BIN env override, PATH, then common install locations. The legacy the original project vendor path is removed entirely — the public package must not default to a personal project layout. Boot fails fast with a clear error when no llama-server binary is found, including a pointer to MBA_LLAMA_SERVER_BIN. The binary check is skipped for ollama, which has no local binary to spawn. Co-authored-by: Cursor <cursoragent@cursor.com>
Replace the personal the original project.dev domain in the adapter apiVersion with mba.ai/v1alpha1. The loader, model catalog, and MCP server loader now accept both the new canonical version and the legacy mba.ai/v1alpha1 for backward compatibility. Draft adapters (mba pull) and family YAMLs now emit mba.ai/v1alpha1. All repo fixtures and tests are updated to the new version, with explicit backward-compat tests for the legacy value. Docs and ADRs (0084, 0091, the manual) are updated to reference the new version and note the legacy alias. Co-authored-by: Cursor <cursoragent@cursor.com>
…ecret - Remove hardcoded apiKeyRef default - Remove hardcoded profile UUID 51cf1714 fallback in sync-endpoints CLI - resolveVsCodeLmConfigPath now returns undefined when VS Code is not detected - Service and CLI skip sync gracefully when config path or MBA_VSCODE_LM_API_KEY_REF is missing - Update tests to use neutral apiKeyRef and cover missing VS Code case Co-authored-by: Cursor <cursoragent@cursor.com>
…public package - Replace user-facing [[mba:]] BCB rule messages with [[mba:]] - Replace mba BCB kill labels with mba - Remove the original project references from chat-message and mba/types comments - Rename llama.cpp fork variant to llama.cpp across types, server, boot, and tests - Replace cyard- temp dir prefixes in tests with mba- - Replace /home/user personal paths in tests and docs with neutral examples - Remove the the original project-specific negative test in server-boot.test.ts Co-authored-by: Cursor <cursoragent@cursor.com>
- Only accept mba.ai/v1alpha1 in core loader, model catalog, and MCP loader - Remove LEGACY_MBA_API_VERSION and SUPPORTED_API_VERSIONS constants - Replace legacy compatibility tests with rejection tests - Update ADR-0084 and .Manual to document the single supported apiVersion Co-authored-by: Cursor <cursoragent@cursor.com>
- New migrate-adapters.ts module scans adapter tree for .yaml files - Rewrites legacy apiVersion lines (e.g. mba.ai/v1alpha1) to the canonical mba.ai/v1alpha1 - Defaults to dry-run; --write required to mutate files - Wired into mba CLI as a local-only command (no service needed) - Tests cover dry-run, write, preservation of other YAML content, and skipping non-legacy files Co-authored-by: Cursor <cursoragent@cursor.com>
- Add 'mba' script to packages/core to run src/cli/mba.ts via tsx - Add root 'mba' script that calls the workspace script so 'npm run mba -- ...' works from the repo root Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
…ne profile persistence Co-authored-by: Cursor <cursoragent@cursor.com>
…clamping overlay - Add service/machine-info.ts to detect CPU, RAM, and GPU (nvidia-smi, system_profiler, lspci, PowerShell) with optional MBA_MACHINE_INFO override. - Add service/machine-store.ts to persist detected specs to ~/.config/mba/mba/machine.json and refresh at boot with a human-readable diff. - Add service/gguf-memory-estimator.ts to estimate llama-server RAM/VRAM needs from GGUF metadata and compute a max-fitting ctxSize. - Add service/machine-overlay.ts to clamp ctxSize, gpuLayers, threads, and parallel to the detected machine (no GPU -> gpuLayers 0, threads/parallel <= cores, ctxSize fits memory budget). - Wire machine profile into the daemon boot path and recipe resolution (recipe-resolution.ts, server-boot.ts, server.ts, main.ts) and the resolve-server-recipe CLI. - Export ResolvedLlamaFlags from mba/server-flags.ts so the overlay can keep the fully-populated flag shape. - Add tests for all new modules. Co-authored-by: Cursor <cursoragent@cursor.com>
- Add machineOverlay setting to global config, persisted to ~/.config/mba/mba/machine-overlay.json, defaulting to enforce. - Add GET/POST /config/machine-overlay endpoints to the service. - Add mba machine-overlay [enforce|warn|off] CLI command. - Wire the mode through recipe resolution and bootServer: - enforce: clamp recipes to fit the machine; refuse boot if it still doesn't fit. - warn: log what would be clamped, but boot the original recipe unchanged. - off: skip the overlay entirely. - Refactor applyMachineOverlay to distinguish originalFits vs clampedFits so warn mode reports whether the original recipe actually fits. - Update tests and /resolve_config /status responses to include the new setting. Co-authored-by: Cursor <cursoragent@cursor.com>
- Fix lspci parsing to extract clean vendor/device names instead of the raw line with bus ID and controller class. - Try nvidia-smi on PATH and common absolute paths (/usr/bin, /usr/local/bin) so the daemon can still find it when the environment differs from the shell. - Use the first GPU with reported VRAM for the machine overlay, instead of always using the first GPU. This prevents an integrated GPU listed before a discrete NVIDIA card from making MBA think no VRAM is available. - Add tests for lspci parsing and multi-GPU VRAM selection. Co-authored-by: Cursor <cursoragent@cursor.com>
- Prefer nvidia-smi --query-gpu=name,memory.total --format=csv,noheader over XML parsing. The CSV format is simpler, more stable, and easier to test. - Keep the XML parser as a fallback in case a future environment returns different CSV formatting. - Add parseNvidiaSmiCsv and tests for single/multi-GPU output. Co-authored-by: Cursor <cursoragent@cursor.com>
- Extend MachineInfo with cpuModel, cpuSpeedMHz, cpuArchitecture, cpuFlags, and cpuPhysicalCores. - Read CPU model/speed from os.cpus(); architecture from os.machine(); physical core count and feature flags from /proc/cpuinfo on Linux. - Update machine-store validation to accept the new optional fields. - Update daemon startup log to print CPU model and physical/logical core count. - Add tests for CPU detection, physical core counting, flag extraction, and validation of malformed cpuFlags. Co-authored-by: Cursor <cursoragent@cursor.com>
- diffMachineInfo was not comparing cpuModel, cpuPhysicalCores, cpuArchitecture, cpuSpeedMHz, or cpuFlags, so a profile that only gained new fields was not rewritten to disk. Add those comparisons and update formatMachineInfo to include the CPU model in the initial-profile message. - Add a test proving that a profile missing CPU fields gets updated when the detector adds them. Co-authored-by: Cursor <cursoragent@cursor.com>
1. Refine gguf-memory-estimator so weights, KV cache, and compute buffer are all split between RAM and VRAM by the gpuLayers / blockCount ratio. This makes partial offloading estimates far more accurate than the old v1 rule that put all KV/cache in VRAM as soon as any layer was offloaded. Add a split breakdown to the returned estimate for diagnostics. 2. Fix machine-overlay to re-check fit after applying non-ctxSize clamps (e.g. gpuLayers forced to 0 because no GPU is available). A recipe that looked like it fit only because 100%% offload shifted memory to VRAM could now fail in RAM, and ctxSize is clamped accordingly. Also mark a recipe as not fitting when it requires VRAM but the machine has no usable GPU. 3. Add mba estimate-memory <model.gguf> as a local-only CLI command. It reads GGUF metadata, prints the RAM/VRAM estimate and split, and compares it against the persisted machine profile, reporting the largest ctxSize that fits. Flags supported: --ctx-size, --gpu-layers, --batch-size, --ubatch-size, --cache-type-k, --cache-type-v, --flash-attn. Tests added for the proportional split and the new CLI command. Co-authored-by: Cursor <cursoragent@cursor.com>
- Add findMaxFittingGpuLayers to the GGUF memory estimator so we can compute the largest GPU-layer count that fits in detected VRAM. - Compute machine-aware hints in model-config.ts for ctxSize, gpuLayers, threads, and parallel from the persisted machine.json profile. These hints show the real bounds imposed by the host: RAM ceiling for ctxSize, VRAM ceiling for gpuLayers, and CPU core count for threads/parallel. - Wire the hints into GET /models/config so the CLI and any other consumer see them automatically. - Update the mba CLI to display both the profile hint and the machine hint in and in the interactive models flow, and to pass the merged hint into the value prompt so users know the safe range before editing. - Add tests covering readModelDials with machine info and the service endpoint response. Co-authored-by: Cursor <cursoragent@cursor.com>
- Replace all ADR 'Deciders: project maintainer + agent' with 'project maintainer + agent'. - Replace real GPU/CPU product names in tests with neutral placeholders: - Example NVIDIA GPU / 3060 / 4090 -> Example GPU / Example High-End GPU - Example CPU Model -> Example CPU Model - Apple M3 Pro -> Apple GPU Example - Intel UHD Graphics -> Intel iGPU Example - NVIDIA GA104/GA102 chip codes -> GPU Chip / GPU Chip 2 - Update ADR-0103 example machine profile to use 'Example NVIDIA GPU'. - Neutralize remaining the original project references in user-facing Manual docs. - Tighten machine-info.ts doc comments to use generic examples. No functional changes; tests updated to match the neutral values. Co-authored-by: Cursor <cursoragent@cursor.com>
- Remove deprecated LEGACY_MBA_* env aliases (MBA_SERVICE_URL, MBA_BASE_DIR are canonical) - Remove ~/.mba base-dir migration and migrateLegacyBaseDir - Remove mba migrate adapters command and legacy apiVersion scaffolding - Replace the original project / the original project / cyard references in ADRs, .Manual, and code comments - Rename test temp prefixes from mba-* to mba-* - Update tests to expect neutral legacy apiVersion rejection Co-authored-by: Cursor <cursoragent@cursor.com>
Record that 0096–0100 and 0103 are accepted, and that ADR-0101 Steps 1–2 are shipped while AMPI Steps 3–5 remain. Co-authored-by: Cursor <cursoragent@cursor.com>
Any escalation tier can summon a recipe; sweep-duplicates asks CM to drop trailing duplicate tool pairs. Default kill behavior is unchanged. Co-authored-by: Cursor <cursoragent@cursor.com>
Recipes name a CM intent; applyCm is the only splice. ADR-0105 records sanitize/assist/sanction/recover and write/mark/sweep/compact. Co-authored-by: Cursor <cursoragent@cursor.com>
Replace the old cut names with replace/insert/set-mark/sweep/compact, keep sweep-duplicates as a sanitize alias, and persist scratch/pin as mba.mark on the tool pair. Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
The old single-file dump stacked pickers and raw argv tokens. Group commands under models/servers/machine, print a grouped flag card before boot, and erase leftover menu frames so each screen stands alone. Co-authored-by: Cursor <cursoragent@cursor.com>
Explain MBA as a local-first per-model daemon (adapter, BCB, AMPI) and put install, onboarding, and CLI after that story instead of a package dump. Co-authored-by: Cursor <cursoragent@cursor.com>
The rewrite kept the work-in-progress sentence but dropped the status badge. Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
…tant files. npm pages cannot reach docs/adr, and Cursor/IDE scratch should stay off the public tree. Project Cursor rules remain tracked. Co-authored-by: Cursor <cursoragent@cursor.com>
Bump core to 0.1.15 and take off ADR's from readMe's
chore(deps-dev): bump vitest from 3.2.7 to 5.0.0
…erver-2.1.1 chore(deps): bump @hono/node-server from 1.19.17 to 2.1.1
…26.4.1 chore(deps-dev): bump @types/node from 22.20.1 to 26.4.1
…heckout-7 chore(deps): bump actions/checkout from 4 to 7
Bumps [hono](https://github.com/honojs/hono) from 4.13.3 to 4.13.7. - [Release notes](https://github.com/honojs/hono/releases) - [Commits](honojs/hono@v4.13.3...v4.13.7) --- updated-dependencies: - dependency-name: hono dependency-version: 4.13.7 dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com>
chore(deps): bump hono from 4.13.3 to 4.13.7
Cursor Co-authored-by lines are left in place. Co-authored-by: Cursor <cursoragent@cursor.com>
…om npm. Co-authored-by: Cursor <cursoragent@cursor.com>
Link the core changelog from the npm README
Bumps [actions/setup-node](https://github.com/actions/setup-node) from 4 to 7. - [Release notes](https://github.com/actions/setup-node/releases) - [Commits](actions/setup-node@v4...v7) --- updated-dependencies: - dependency-name: actions/setup-node dependency-version: '7' dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com>
Bumps [tsx](https://github.com/privatenumber/tsx) from 4.23.12 to 4.23.13. - [Release notes](https://github.com/privatenumber/tsx/releases) - [Changelog](https://github.com/privatenumber/tsx/blob/master/release.config.cjs) - [Commits](privatenumber/tsx@v4.23.12...v4.23.13) --- updated-dependencies: - dependency-name: tsx dependency-version: 4.23.13 dependency-type: direct:development update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com>
Bumps the npm_and_yarn group with 1 update in the / directory: [qs](https://github.com/ljharb/qs). Updates `qs` from 6.15.3 to 6.16.0 - [Changelog](https://github.com/ljharb/qs/blob/main/CHANGELOG.md) - [Commits](ljharb/qs@v6.15.3...v6.16.0) --- updated-dependencies: - dependency-name: qs dependency-version: 6.16.0 dependency-type: indirect dependency-group: npm_and_yarn ... Signed-off-by: dependabot[bot] <support@github.com>
…rn-049d63b98d chore(deps): bump qs from 6.15.3 to 6.16.0 in the npm_and_yarn group across 1 directory
Recreate Dependabot bumps: setup-node v7 and tsx 4.23.13
Co-authored-by: Cursor <cursoragent@cursor.com>
Clear CodeQL ReDoS alerts in version parse and URL trim
Bumps [typescript](https://github.com/microsoft/TypeScript) from 5.9.3 to 7.0.2. - [Release notes](https://github.com/microsoft/TypeScript/releases) - [Commits](microsoft/TypeScript@v5.9.3...v7.0.2) --- updated-dependencies: - dependency-name: typescript dependency-version: 7.0.2 dependency-type: direct:development update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com>
dependabot
Bot
added
dependencies
Contributor
Author
|
OK, I won't notify you again about this release, but will get in touch when a new version is available. If you'd rather skip all updates until the next major or minor version, let me know by commenting If you change your mind, just re-open this PR and I'll resolve any conflicts on it. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Bumps typescript from 5.9.3 to 7.0.2.
Release notes
Sourced from typescript's releases.
Commits
1e4744dMerge branch 'main' into ts7-releasea5a219cmicrosoft/typescript-go#4558ecfe30dUpdate status localization5de25b5Hide executable name in TypeScript statusd7ce74aShow bundled TypeScript version for packaged servers29be66aCorrect TS 7 release version to 7.0.2ed2bd1bMerge branch 'main' into ts7-release8873075Bump the github-actions group across 1 directory with 3 updates (microsoft/ty...9427131Set up stable / nightly extension split, other prep (microsoft/typescript-go#...d4eaca5microsoft/typescript-go#4549Maintainer changes
This version was pushed to npm by microsoft1es, a new releaser for typescript since your current version.
Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting
@dependabot rebase.Dependabot commands and options
You can trigger Dependabot actions by commenting on this PR:
@dependabot rebasewill rebase this PR@dependabot recreatewill recreate this PR, overwriting any edits that have been made to it@dependabot show <dependency name> ignore conditionswill show all of the ignore conditions of the specified dependency@dependabot ignore this major versionwill close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this minor versionwill close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this dependencywill close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)