* test(hwfit): assert the Apple matcher, not the general lookup, in the non-Apple guard
f7aa2de (#2564) added test_non_apple_gpu_with_cores_does_not_match, which
asserts _lookup_bandwidth(RTX 4090) is None. But '4090': 1008 has been in
the general GPU_BANDWIDTH table since v1.0, so _lookup_bandwidth correctly
returns the card's real bandwidth and the test fails (expected None, got
1008) - reddening the required pytest gate on dev and, by inheritance,
every open PR.
The guard's actual intent is that the Apple-specific bandwidth path does
not false-match a non-Apple card that carries a gpu_cores count. Point
the two asserts at _lookup_apple_bandwidth, which returns None for any
name without 'apple' regardless of the general table. The general-lookup
behavior (4090 -> 1008) is correct and untouched.
* fix(hwfit): route string GPU names through the Apple bandwidth helper
Second half of the #2564 regression (RaresKeY review on #4303). That
change moved the Apple tiers out of the generic GPU_BANDWIDTH table into
the dict-only _lookup_apple_bandwidth, but _lookup_bandwidth only called
that helper for dict inputs. A bare-string caller like
_lookup_bandwidth("Apple M3 Max") therefore fell through to the generic
table, found no Apple key, and returned None instead of the conservative
tier. Route both dict and string inputs through the Apple helper (a
string carries no gpu_cores, so it gets the model's lowest tier).
Regression added for the string path plus a non-Apple string control.
* fix: resolve Apple Silicon bandwidth variants
* fix(hwfit): preserve string lookup path in _lookup_bandwidth
* fix(hwfit): guard Apple bandwidth lookup against false GPU matches
Add "apple" not in gn check to _lookup_apple_bandwidth() so that
non-Apple GPUs with "m3"/"m4"/"m5" in their names (e.g. NVIDIA
Quadro M4 000) don't incorrectly match Apple bandwidth tiers.
Addresses @o3LL review comment on PR #2564.
* Agent: make skill-prescribed tools actually callable
The skill index and matched-skill procedures are injected into the
prompt, but tool selection never followed: manage_skills wasn't in the
RAG-selected schema list (so the model substituted manage_memory), and
a matched skill could prescribe tools (grep, read_file) the model had
no schema for. Now:
- manage_skills rides along whenever the owner has any skills indexed
- a Jaccard-matched skill's requires_toolsets join the selection
- viewing a skill mid-turn via manage_skills unlocks its
requires_toolsets for subsequent rounds
- admin-intent turns send _ADMIN_TOOLS schemas, matching the prompt
text _build_base_prompt already advertises
- index_for(active_toolsets=None) no longer hides requires_toolsets
skills from callers that don't know the active set
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* Agent: validate skill requires_toolsets against known tools, not TOOL_SECTIONS
grep/glob/ls ship as function schemas without a prompt-prose section,
so gating on TOOL_SECTIONS silently dropped them from a skill's
requires_toolsets.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
Make src.youtube_handler a compatibility wrapper around services.youtube.youtube_handler so transcript state, URL parsing, and timeout behavior no longer diverge.
raw.githubusercontent.com serves Markdown as text/plain, JSON APIs and raw
config files serve application/json, and a lot of code and tool documentation
lives in .md/.txt. fetch_webpage_content only handled PDF and HTML, so a
non-HTML body produced empty content and web_fetch reported 'no readable text
content'. Add a branch that returns the body verbatim for non-HTML text/*,
JSON (application/json and +json), and a .md/.txt/.text/.json URL-suffix
fallback for mislabeled octet-stream. HTML and PDF handling unchanged.
Fixes#3808
* Switch to ddgs
duckduckgo_search was deprecated, this is the recommended replacement
* Update test_service_search_provider_guards.py
According to review comment
* fixed confusing credentials prompt
* fix(setup): return status from create_default_admin function
* fix(setup): initialize admin creation status in main function
* fix(setup): enhance admin creation feedback and status handling
* Enhance admin user login messages with conditional feedback based on creation status
* Refine admin user creation feedback messages for clarity and actionability and formatted code
* Add fallback error message for admin creation failure in setup script
* Add run script for Uvicorn with dotenv integration
* Refactor server runner to use argparse for host and port configuration
* Remove captured output print statement from server runner
* Fix server runner to ensure cross-platform compatibility and improve log handling
* removed run.py to match original repo
* Fixing custom search not working properly
* Refactor search settings event listeners for improved functionality and clarity
* Update search function signatures to use Optional for count parameter
* revert changes
* fixed broken merge issue
* Delete services/chat_data_scraper.py
added by mistake
---------
Co-authored-by: Alexandre Teixeira <111787685+alteixeira20@users.noreply.github.com>
Surface a lot of accumulated cookbook + UI work as a single non-agent
commit so the agent rework lands cleanly.
Highlights:
- Ollama as a first-class backend in the Cookbook:
* Download input accepts ollama-style names (name:tag) → backend=ollama
* /api/cookbook/ollama/library (cached scrape of ollama.com + curated
fallback so classic models like qwen2.5 stay reachable)
* "Browse Ollama library" toggle below Download with size chips
* Engine=Ollama in hwfit toolbar merges the Ollama library into the
main scan list as per-tag rows with the same Fit/Param/Quant/VRAM
columns; click → fills Download input
- API Tokens form added to Integrations panel (matching wired
loadTokens()/initTokenForm() that had no HTML)
- Serve panel polish: Advanced fold tightening (-8px nudges on vLLM
checks, Extra args, Spec row), n_cpu_moe + Split Mode controls
pulled up 8px to align with the row's checkboxes, GGUF File dropdown
exposed for Ollama backend, GPU re-render on Edit serve restore,
_forceBackend flag so saved serveState wins over backend detection,
cookbook:servers-changed CustomEvent so panels don't need refresh
- Models page redesign: Add Models row (URL + hidden API key reveal +
Type select + Scan/Ollama/Key/Test/Add icon buttons), Probe All +
Clear-offline buttons in Added Models toolbar, offline-pill removed
(opacity already conveys state), Engine dropdown gains Ollama option
- _ping_endpoint probes /v1/models then base, accepts 4xx as
reachable (vLLM returns 404 on bare /v1, fully working endpoints
were showing offline)
- Diagnosis card: × dismiss + Copy bundle buttons restored on the
serve error feedback card
- Orphan tmux sweep re-enabled behind a 60s rate-limit + background
Thread (off the main event loop) so dead serves get discovered
- cookbook_routes auto-register watchdog: drops the endpoint if the
serve session exits non-zero within the first ~3min
- ollama-rocm sidecar awareness in download wrapper (`docker exec
ollama-rocm ollama pull` when host ollama isn't installed)
- Skill extractor sets initial_status="published" when
auto_approve_skills pref is on (audit demotes later)
- Skill list / model list / cookbook scan misc polish
* fix(memory): auto-memory extracted nothing — flatten window so the prompt ends on a user turn
extract_and_store appended the recent window as raw alternating role messages
after the system prompt. Since the window is the last N messages, the prompt
usually ENDED on an assistant turn — and a chat model given a prompt ending on
an assistant turn returns an empty completion (nothing to answer). The result
was facts=[] → "Auto memory extraction ran: 0 candidates" on every run, so no
memories were ever stored, while skill extraction (which flattens the transcript
into a single user message) worked fine.
Flatten the window into one user message ending with an explicit instruction,
mirroring the skill extractor, so the model always responds. Also harden parsing
for reasoning models, matching the audit path which already does this:
- raise max_tokens 500 → 4096 (a reasoning model spends the budget on <think>
before emitting JSON; 500 truncated it before any JSON appeared);
- strip <think>/prose preambles via strip_think and slice the embedded JSON
array before json.loads, instead of bombing on char 0.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* chore: tighten memory-extraction-empty-completion — clarify JSON-slice comment re prior strip steps
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* docs(memory): reframe the comment to the accurate root cause (raw-chat framing)
The earlier comment leaned on "ends on an assistant turn -> empty completion",
which is only one failure mode. The dominant cause, confirmed by a controlled
repro (0/6 old vs 6/6 new on this model), is that passing the window as raw chat
messages makes the model treat it as a conversation to continue rather than a
transcript to analyze, so it returns [] even when durable facts are present.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* test(memory): cover extraction JSON parsing + slice trailing commentary unconditionally
Factor the strip/fence/slice/json.loads logic out of extract_and_store into
a pure module-level helper _parse_extraction_json(raw) -> list and drop the
'text[0] != "["' guard so the array is sliced whenever both brackets exist
(fixes trailing commentary like '[...] Done!' reaching json.loads).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
* refactor(constants): single source of truth for data dir + merge core/src constants
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* docs(contributing): use named src.constants for data paths, drop core/constants references
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
raise_for_status() raises httpx.HTTPStatusError for 4xx/5xx responses,
but the surrounding try/except only caught httpx.RequestError (network
errors) and RateLimitError (429). Any other HTTP error code propagated
uncaught up through chat_processor -> chat_helpers -> chat_routes and
surfaced as a 500 Internal Server Error.
Added an explicit except httpx.HTTPStatusError clause that logs a warning
and returns an empty result, matching the behaviour already in place for
network errors.
Also adds focused regression tests that exercise the real
fetch_webpage_content() path with a mocked _get_public_url:
- 403/404 responses return the standard empty-result shape instead of
raising, proving the new HTTPStatusError handling works end to end.
- 429 responses still take their own dedicated rate-limit branch (the
status_code == 429 check runs before raise_for_status() is reached),
keeping that behaviour distinct from the new generic HTTPStatusError
handling.
Dropped the unrelated builtin_mcp.py change that had been carried over
from a rebase; that fix is tracked separately in #2018 and this branch
should stay scoped to the search content fetch path.
Closes#2148
* fix(platform): add WSL compatibility functions and path translation
fix(cookbook): enhance model scan script to support additional HuggingFace cache paths
fix(hardware): improve cache key generation for remote SSH context
test(tests): add tests for WSL detection and path translation functionality
* fix(cookbook): prefer prebuilt wheels for llama-cpp-python and normalize package aliases
* fix: enable StrictHostKeyChecking in nvidia probe
refactor: consolidate ssh & powershell command execution to utility functions in core module
refactor: consolidate nvidia path candidates in to single variables in core module
tests: add tests for new utility functions
* fix: correct wrong variable name
* fix(skill-extractor): walk all brace candidates so stray braces in prose do not swallow valid JSON
The extractor sliced from the FIRST brace to the LAST brace to recover
JSON embedded in surrounding commentary. When the model emits stray
braces before the JSON object, the slice produces invalid JSON,
json.loads raises, and the exception is swallowed -- the skill is
silently lost.
Fix: walk each brace candidate left-to-right and attempt json.loads on
each slice. The first candidate that parses successfully wins. If none
parse, json.loads on the original text raises and the existing
JSONDecodeError handler logs and returns None as before.
Tested locally -- 8/8 tests passed:
tests/test_extract_skill_json_nonstring.py (2 passed)
tests/test_skill_extractor_rows.py (1 passed)
tests/test_search_content_extraction_parity.py (2 passed)
tests/test_deep_research_search_error.py (3 passed)
Closes#2199
* test(skill-extractor): add focused repro for stray-brace JSON recovery
* test(skill-extractor): add regression test for leading invalid-brace fragment
Addresses the remaining edge case from review: a response that *starts*
with a brace but the leading fragment isn't valid JSON (e.g.
'{not json}\n{"title": "Valid later", ...}') still needs to recover
the valid skill object that follows.
_extract_json_object (already on dev) handles this correctly — it tries
the whole de-fenced string first, then walks each '{' candidate left-to-
right regardless of whether the response begins with '{', so the leading
invalid fragment no longer short-circuits recovery of the real object.
Updates the comment at the call site to call this out explicitly and adds
a regression test covering exactly the scenario described in review.
services/search/cache.py set CACHE_DIR = services/cache (the source tree) and
mkdir'd it at import, unguarded. In Docker services/ is the read-only image
layer, so the mkdir fails at import (same class as the analytics bug #2366).
Move the cache under DATA_DIR/cache (writable on Docker and native) and wrap
the mkdir so an unwritable path disables disk cache instead of crashing import.
Part of #3331.
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
* fix: move search analytics log to writable /app/logs volume
services/search/analytics.py opened a FileHandler at module import
time pointing to /app/services/search_engine_error.log — inside the
container image's read-only layer. The process runs as non-root so
the open() fails with PermissionError, crashing uvicorn before it
ever binds. ANALYTICS_FILE had the same problem.
Both paths now point to /app/logs (bind-mounted from the host data
directory). The FileHandler creation is wrapped in try/except so a
missing mount doesn't hard-crash on import.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix: derive log dir from DATA_DIR instead of hardcoded /app/logs
Fixes reviewer feedback on #2366: /app/logs only exists inside Docker,
so native runs couldn't write the analytics file. DATA_DIR resolves to
the repo's data/ directory on native and /app/data (writable mount) in
Docker, making both the error log handler and ANALYTICS_FILE work on
every platform.
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
The nvidia-smi absolute-path fallback in _detect_nvidia() was gated
on _remote_host, so it never ran for local detection. On systems
where nvidia-smi is not in the default PATH (e.g. WSL: /usr/lib/wsl/lib/),
this caused the Cookbook to report 'No GPU' even when nvidia-smi works
from an interactive shell.
Two issues fixed:
1. Removed the _remote_host gate so the absolute-path scan runs for
local detection too.
2. For local execution, pass arguments as a list instead of a string
so subprocess.run() resolves the absolute path correctly. Remote
(SSH) execution keeps the string form, which the SSH command builder
handles.
Co-authored-by: Bipin Mishra <bipin.mishra@atlascopco.com>
maybe_extract_skill() sliced the LLM response from the first '{' to the
last '}'. When a model emits a stray brace in prose before the real
object (e.g. "uses {placeholder} then {...}"), the slice starts at the
prose brace, json.loads fails, and a valid skill is silently dropped.
Factor parsing into _extract_json_object(), which tries the whole
(de-fenced) string first and then each '{' start position, returning the
first candidate that parses to a JSON object.
Adds tests/test_skill_extractor_json.py.
_fallback_memory_candidates matched both positive (prefer/like/love) and
negative (hate / do not like / don't like) sentiment verbs in one regex
alternation, then formatted every hit as "User prefers {X}.". So
"I hate cilantro" was stored as "User prefers cilantro." -- the inverse of
what the user said. These fallback facts are persisted to memory and later
re-injected into the model's context, so the inverted preference actively
misleads the assistant.
Capture the matched verb and branch on it: negatives become
"User dislikes {X}.", positives stay "User prefers {X}." (still filed under
the existing "preference" category).
Supported by Claude Opus 4.8
Co-authored-by: SurprisedDuck <288741682+SurprisedDuck@users.noreply.github.com>
* Convert to different style of comment to make it easier to work with, fix formatting inside Powershell script.
* Grab VRAM amount from driver's registry keys.
* Fixed regression on NVIDIA GPUs
This change extends both the `PATH` variable and the list of absolute paths used to locate the `nvidia-smi` package to include `/usr/lib/wsl/lib`.
This path is a candidate for the default location of nvidia-smi for WSL machines (tested on WSL Ubuntu 22.04.5).
* feat(skills): import SKILL.md bundles from public GitHub URLs
Supports GitHub tree/blob/raw links and skills.sh pages that resolve to GitHub.
Installs SKILL.md plus sibling text assets under data/skills/imported/.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(skills): admin-gate URL import and validate redirect hosts
- require_admin on POST /api/skills/import-from-url (matches other skill admin routes)
- reject cross-host redirects after httpx follow_redirects
- test for redirect host validation
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(skills): match Brain Add panel import/submit button styles
- Skill URL Import: theme-io-btn + download icon (same as memory Import)
- Add Skill submit: confirm-btn confirm-btn-primary
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(skills): allow api.github.com during directory import
Real imports hit the GitHub contents API after redirects; whitelist
api.github.com and add regression tests. Shrink Import button with flex:none.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(skills): align skill Import button with URL input row
Match memory-add-input height (28px) in memory-add-row and center the
download icon with flexbox instead of vertical-align hacks.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(skills): cancel modal-body margin on skill Import button
The skill Import button sits in .memory-add-row beside an input; the
global .modal-body button { margin-top: 6px } rule only affected buttons,
pushing Import down and misaligning the download icon. Reset margin-top
and match Memory Import SVG markup at 28px row height.
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(skills): surface GitHub API errors on URL import
Pass through GitHub response messages (especially 403 rate limits) as
SkillImportError instead of a generic download failure.
Co-authored-by: Cursor <cursoragent@cursor.com>
---------
Co-authored-by: Cursor <cursoragent@cursor.com>
#1473 converted the title and sports-hint matches in services/search/ranking.py
to word boundaries but left two raw substring tests:
- snippet_score: 'term in snippet.lower()' — query term 'port' hits
'transport'/'support', inflating a result's relevance.
- news_quality_adjustment: 't in text or t in netloc' for the subject term —
query 'us' substring-matches 'business'/'music', so an off-topic page
wrongly escapes the off-topic penalty on a country/subject news query.
Add a _has_word helper (the same \b...\b pattern title_score already used) and
route all three word checks (title, snippet, subject) through it, so the file
stays consistent and a future partial fix can't reintroduce the same bug class.
Pure ranking refinement: scores change only for spurious substring matches; no
API or schema change.
(cherry picked from commit 22bd23f044)
Co-authored-by: ghreprimand <203024559+ghreprimand@users.noreply.github.com>
extract_and_store dedups each extracted fact against the vector store
before the (owner-scoped) text fallback. The vector store is a single
shared ChromaDB collection storing only {"source": "memory"} — no
owner — and find_similar queries it with no owner filter, so it can
return a memory_id belonging to a different tenant. The old code
continue'd (skipped storing) on any vector hit without checking
ownership, so when ChromaDB is healthy (the common path) a user's
freshly-extracted fact was silently dropped because it was merely
semantically similar to another user's memory — the text fallback that
IS owner-scoped never ran. Gate the skip on the matched memory being
this user's own (or legacy unowned), mirroring the text dedup predicate;
cross-tenant or stale matches fall through. Same bug class as #1743.
Odysseus only supports llama.cpp on Windows (vLLM/SGLang are
explicitly blocked). llama.cpp requires GGUF, so AWQ/GPTQ/FP8
safetensors models without a GGUF alternate should not be
recommended in the Cookbook on Windows hosts.
Changes:
- hardware.py: add 'platform': 'windows' to _detect_windows()
so downstream logic can identify Windows hosts.
- fit.py: include is_windows in the existing GGUF-only filter
alongside apple_silicon and consumer_amd.
- tests: add test_hwfit_windows.py with regression tests.
Fixes#122, #614 (root cause: unservable models recommended).
The multi-GPU GGUF filter at fit.py:380 returned None unconditionally
for Q*/IQ quants on 2+ GPU systems. When the caller explicitly passes
target_quant, they are asking 'what happens if I try this?' and expect
a structured no_fit response, not a silent None.
Fix: skip the filter when target_quant is explicitly provided so the
call falls through to the existing no_fit path.
Fixes #
Backend (services/hwfit + routes):
- VRAM column sort now shows global highest first (was special-cased to
ascending then truncated top-N, which made "highest VRAM" mathematically
unreachable). Every column path uses reverse=True for the truncation.
- Hardware probe cache TTL 30min -> 24h so changing filters doesn't keep
re-probing the rig during a session; Rescan button still forces fresh.
- Multi-GPU rigs filter GGUF Q*/IQ quants (vLLM/SGLang can't serve them);
default non-prequantized to BF16 on 2+ GPUs.
- AWQ / AWQ-8bit / GPTQ-8bit get a -1.0 quality penalty so FP8 wins ties.
- Version-aware tiebreaker (parse Mn.n / Vn) — MiniMax-M2.7 ranks above M2.5.
- hf_models.json: zai-org/GLM-5.1 added; zai-org/GLM-5 quantization flipped
Q4_K_M -> BF16. DeepSeek-V4-Flash / -Pro + their -Base variants registered
with new FP4-MoE-Mixed / FP8-Mixed quant keys (calibrated BPP from the
actual 156 GB / 284 GB disk footprints).
- New FP4-MoE-Mixed + FP8-Mixed entries in QUANT_BPP / QUANT_SPEED_MULT /
QUANT_QUALITY_PENALTY / QUANT_BYTES_PER_PARAM / PREQUANTIZED_PREFIXES.
Frontend — Scan/Download:
- Engine + Quant swapped in the toolbar; Quant defaults to "All".
- Ctx (range slider) ported from origin/main: 8k/16k/32k/50k/128k/Max. Drag
re-sorts by vram ascending (smallest fitting first); back to Max → score.
- Ctx slider rail now visible — was background:transparent in a duplicate
later-cascade rule. Hardcoded grey + !important.
- Search input moved to the far right of the toolbar.
- Type/Standard default; "Context" not uppercased; Search placeholder dimmed.
- Engine "?" + Quant "?" inline help chips inside their dropdown boxes.
- Fit-column dot toggles fit-only filter; un-toggling re-sorts by VRAM desc.
- Quant column truncates to 9 chars + ellipsis ("FP4-MoE-M..."), full in
tooltip. Smart title-suffix strips the parts already in the repo name
(QuantTrio/MiniMax-M2-AWQ + quant AWQ-4bit -> just "(4bit)").
- Conditional warning for safetensors models on non-GPU rigs only.
- Dependency Install / Installed / Installed▾ / N/A all 75.85px wide.
- Rebuild llama.cpp moved into the llama_cpp dep row, styled as a tag.
- Foldable Download admin-card (h2 chevron); line under h2 only when folded.
- HF token save gets a green ✓ + "Saved" flash.
- Cached scan no longer counts stalled rows as downloaded.
- Footer: "Request it →" link with GitHub mark to the public discussion
(#1962) for model-add requests.
Frontend — Running tab:
- Strict download-finish check (DOWNLOAD_OK or /snapshots/, not bare
"Download complete"). True overall % for multi-shard downloads:
((N-1)+frac)/total instead of hf_transfer's per-shard aggregate.
- ETA in the uptime ticker: "downloading: 12m 34s · ETA 1h 23m".
- Clear button kills the tmux session too; if the output still shows a
live shard line, the pill is hidden + relabels as "reconnect" + revives
on click.
- Self-heal: on cookbook open AND every bg-monitor cycle (10s, throttled
to 8s), scan persisted done/error/crashed downloads and probe their
tmux session — if alive, flip status back to running and reattach.
- Per-launch zombie probe: clicking Download on a model whose persisted
state is done but tmux is still alive revives the existing task and
refuses to start a duplicate.
- Pre-launch GPU probe: vllm / sglang / diffusers serve check
/api/cookbook/gpus first; warns + confirms if no GPU is visible.
- Server-side state guard: rejects "done" POSTs for downloads lacking
DOWNLOAD_OK / DOWNLOAD_FAILED / /snapshots/ when the last-mentioned
shard is N<total — stale tabs can't poison persisted state any more.
- Running count includes tasks whose output looks active even if persisted
status got stuck. Dir text on the running row, font matched to uptime.
Serve panel:
- Ctx text input always resets to model max on open (default 20000 when
metadata is missing).
- Max Seqs default 8 -> 4. KV Cache dtype select 32px tall.
- Lightning icon on Launch (same as Action toggle).
- Diagnosis card simplified (no fold/copy/dismiss), suggestion font
matches body; action buttons get icons on the left (Retry/Copy/Edit/
Install/Kill/Switch/etc.).
- Incomplete-download serve warning when model status is
downloading / stalled / has_incomplete.
- MTP "?" tooltip ("supported on a few model families … up to ~3× faster").
Backend (services/hwfit + routes):
- rank_models picks visible set by REQUESTED column, not always score —
sorting by Param now shows highest-param models PERIOD (incl. too_tight).
- New fit_only param. Multi-GPU rigs filter GGUF Q*/IQ quants (vLLM/SGLang
cannot serve them); default non-prequantized to BF16 on 2+ GPUs.
- AWQ / GPTQ-8bit get a -1.0 quality penalty (was 0.0, tied with FP8), so
FP8 wins when both fit.
- Version-aware tiebreaker (parse Mn.n / Vn) — MiniMax-M2.7 ranks above
M2.5 on equal composite score; >=100B integers not misread as versions.
- /api/cookbook/hf-latest no longer drops models without an "NB" pattern in
the repo id (MiniMax-M2.7, DeepSeek-V4-Pro etc. were silently filtered).
- Cached-model scan: atexit flushes models JSON even if the script is
killed mid-walk; each scan_dir wrapped in try/except; timeout 60s -> 180s.
- KB granularity for sub-MB sizes (was "0 MB" for 12 KB shells). New
"stalled" status for shells <1 MB with no .incomplete files.
- /api/cookbook/state POST guard: rejects "done" download tasks lacking
DOWNLOAD_OK / DOWNLOAD_FAILED / /snapshots/ when the last-mentioned
shard is N<total — stops stale tabs from poisoning persisted state.
- hf_models.json: add zai-org/GLM-5.1; flip zai-org/GLM-5 quantization
Q4_K_M -> BF16 (it is the native base, not a quant).
Frontend (static/js):
- Scan/Download toolbar: quant defaults to All; ctx slider (8k/16k/32k/
50k/128k/Max) ported from origin/main with sort=fit on drag, sort=score
on Max. GPU toggle commits _activeCount to maxGpu on initial render. Fit
column header tagged with active budget (RAM / GPU / N GPU).
- Foldable Download admin-card: the Download h2 is the chevron trigger;
state persists in localStorage.
- Download card surfaces destination dir (Dir: <path>). Same dir on running
task row, font/color matched to uptime (9px Fira Code muted, opacity .4).
- Serve panel ctx text input always resets to model max on open. Sub-MB
cached models show with red "download stalled" badge.
- Bulk-select Cancel + Delete reset the Select button label on exit.
- Cookbook running: false-finished bug fixed — DOWNLOAD_OK or /snapshots/
required; bare "Download complete" no longer marks the task done after
the first config file. Clear button now sends tmux kill-session too.
True overall % for multi-shard downloads: ((N-1)+frac)/total instead of
hf_transfer per-shard aggregate.
- Diagnosis card simplified: removed fold toggle, copy button, dismiss X.
Suggestion font matches message body (12px).
- HF token field flashes green check + "Saved" on save.
- Cached scan no longer counts stalled rows as downloaded in Scan/Download.
CSS:
- dep Install button width pinned to 76px to match Installed split.
- task-sub row +1px; task-status badge gets margin-right 8px.
- Ctx slider styled like gallery editor sliders (thin pill rail, red thumb).
- Bulk-select cancel button top -3px -> -5px.
tavily_search, serper_search and google_pse_search parsed response.json()
inside the network try block, which only caught httpx.RequestError and
RateLimitError. When a provider returned a non-JSON body (an HTML error page, a
truncated/empty body, a gateway 5xx), response.json() raised an UNCAUGHT
json.JSONDecodeError that aborted the search in the background — exactly the
'search engines other than SearXNG fail in the background' symptom.
brave_search already handles this correctly: it parses JSON in its own try
block and returns [] on json.JSONDecodeError. Mirror that in the other three
providers so a malformed provider response degrades to no-results instead of
propagating an exception.
Adds tests/test_search_provider_json.py: a non-JSON 200 body now yields [] for
tavily, serper, google_pse, and brave (the last guards the reference behaviour).
Co-authored-by: NubsCarson <nubs@nubs.site>
* fix: match skill tags as whole tokens, not substrings, in retrieval
* test: skill tag matching uses whole tokens, not substrings
* test: give skill fixtures status=published so they reach the scoring path
synthesize() and get_stats() parsed the stored tts_speed with a bare
float(settings.get("tts_speed", "1")). The manage_settings agent tool maps
"speech speed"/"voice speed" to tts_speed and, because the setting's default is
a string, writes the value through unvalidated — so an agent (or a hand-edited
settings.json) can store "fast" or "". After that, GET /api/tts/stats and POST
/api/tts/synthesize both 500 with ValueError until the JSON is corrected by hand.
Parse defensively via a _safe_speed() helper (non-numeric/empty/<=0 -> 1.0),
mirroring the settings layer's tolerance of corrupt config.
Adds tests/test_tts_speed_malformed.py (stats + synthesize) — both raise
ValueError before this change and pass after.
Kenny Van de Maele