@fffangchenghao/dsh-web-search 0.11.0 → 0.12.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/CHANGELOG.md CHANGED
@@ -1,5 +1,33 @@
1
1
  # Changelog
2
2
 
3
+ ## 0.12.0
4
+
5
+ ### Added
6
+ - **Fetched pages come back as article Markdown**, not raw markup. Extraction
7
+ picks the article root, prunes page furniture, renders headings, lists, code
8
+ fences, tables, entities and absolute links, and cuts at `fetchMaxLength` with
9
+ a notice naming the original length. `fetchFormat` can return plain text.
10
+ - **`search_many`** answers one to five queries in a single call with a
11
+ concurrency cap of three and per-query error isolation.
12
+ - **The cache survives a restart.** A versioned snapshot keeps summaries only,
13
+ is written atomically, caps itself at 4 MiB with LRU eviction, and treats a
14
+ corrupt or unreadable file as an empty cache. `cachePersist` and `cacheFile`
15
+ control it.
16
+
17
+ ### Fixed
18
+ - `cacheTtlMs`, `cacheEntries`, `cachePersist` and `cacheFile` are read live, so
19
+ a Settings edit takes effect without a remount, like every other volatile
20
+ field.
21
+
22
+ - `fetchFormat` / `fetchMaxLength` are resolved per request, so the Settings page
23
+ no longer needs a reload for them either; only the cache file path and the
24
+ persist switch are read at mount.
25
+
26
+ ### Known limits
27
+ - The host fetch request carries only a url, so a page cannot be resumed at an
28
+ offset: raise `fetchMaxLength` or fetch a narrower page.
29
+ - Two processes writing the same cache file is last-writer-wins.
30
+
3
31
  ## 0.11.0
4
32
 
5
33
  ### Added
package/README.md CHANGED
@@ -5,11 +5,11 @@
5
5
  - [中文说明](README.zh-CN.md)
6
6
  - [Changelog](CHANGELOG.md) · [Why a plugin's settings go missing: the four-layer chain](docs/plugin-settings-form.md)
7
7
 
8
- > Version 0.11.0. What follows starts with what this means for you; the technical detail is collected in the appendix.
8
+ > Version 0.12.0. What follows starts with what this means for you; the technical detail is collected in the appendix.
9
9
 
10
10
  ## In one sentence
11
11
 
12
- With this plugin installed, you can ask something like "how do I configure the version released last week?" and the model can call `web_search` against public search engines by itself, then work from a ranked list of titles, links and snippets. When a snippet is not enough, it can call `web_fetch` on one of those links and pull the page text into the conversation. When the question is about an npm package, a GitHub repository, a paper or a Q&A thread, it can call `domain_search` instead and ask those dedicated sources directly.
12
+ With this plugin installed, you can ask something like "how do I configure the version released last week?" and the model can call `web_search` against public search engines by itself, then work from a ranked list of titles, links and snippets. When a snippet is not enough, it can call `web_fetch` on one of those links and get the page back as clean Markdown instead of a wall of HTML. When one question needs several separate searches, it can call `search_many` and get each query's results separately. When the question is about an npm package, a GitHub repository, a paper or a Q&A thread, it can call `domain_search` instead and ask those dedicated sources directly.
13
13
 
14
14
  Three things change for you:
15
15
 
@@ -17,7 +17,7 @@ Three things change for you:
17
17
  - When a question is about a package, a repository or a paper, the model can ask the matching source directly instead of hoping a web search understands.
18
18
  - You do not need a search API key, and you do not pay for search separately.
19
19
 
20
- Under the hood it is one sentence too: the plugin registers two providers with the harness's `web` service — `bing` for search and `http-proxy` for fetching pages — and adds one more tool, `domain_search`, for packages, repositories, papers, Q&A and reference entries. These names come up again below.
20
+ Under the hood it is one sentence too: the plugin registers two providers with the harness's `web` service — `bing` for search and `http-proxy` for fetching pages — and adds two more tools: `domain_search` for packages, repositories, papers, Q&A and reference entries, and `search_many` for several searches in one call. These names come up again below.
21
21
 
22
22
  ## The problem it solves
23
23
 
@@ -75,12 +75,12 @@ A **tool** is a fixed capability handed to the agent: search, read a page, run a
75
75
  This is the pair people mix up most. They solve different problems:
76
76
 
77
77
  - **`web_search` finds URLs.** Give it keywords; it returns a list with titles, links and a line or two of snippet.
78
- - **`web_fetch` reads a page.** Give it one URL; it brings back that page's content.
78
+ - **`web_fetch` reads a page.** Give it one URL; it brings back that page's content, cleaned into readable Markdown.
79
79
 
80
80
  | | `web_search` | `web_fetch` |
81
81
  | --- | --- | --- |
82
82
  | You give it | A few keywords | One URL (`http://` or `https://`) |
83
- | It gives you | Eight candidates by default: title, link, snippet (fewer with a time window) | The page body (HTML or plain text) |
83
+ | It gives you | Eight candidates by default: title, link, snippet (fewer with a time window) | The page body, cleaned into Markdown (or plain text) |
84
84
  | Can it answer the question | Only from snippets, often not enough | Yes — the text is right there |
85
85
  | When to use it | You don't know where to look, or want several sources | You know which page to read |
86
86
  | Which provider | `bing` (this plugin) | `http-proxy` (this plugin) |
@@ -94,6 +94,8 @@ In everyday terms:
94
94
 
95
95
  You need both. Search without fetch leaves the model reading blurbs forever. Fetch without search requires a URL — and the model usually does not have one.
96
96
 
97
+ What fetch hands back is the **article**, not the page's furniture: navigation, ads, scripts, footers and forms are stripped, and headings, lists, code and tables survive as Markdown.
98
+
97
99
  ### Dedicated search (domain_search): sources, not pages
98
100
 
99
101
  `web_search` asks "what is on the open web". `domain_search` asks something narrower: "does this package exist, is this repository maintained, is there a paper on this, has someone answered this?"
@@ -112,6 +114,8 @@ Analogy: `web_search` is asking passers-by where the hardware shop is and gettin
112
114
 
113
115
  They team up well: use `domain_search` to find the package or paper, then `web_fetch` to open its homepage or README and read the detail.
114
116
 
117
+ One more helper sits beside them: **`search_many`**. When one question needs several distinct searches — one per candidate library, say — it runs one to five queries in a single call and keeps each query's results apart, so one failed query cannot empty the batch.
118
+
115
119
  ### DeepSeek messages and sessions
116
120
 
117
121
  The model has no memory of its own. On every request, the harness resends the **whole sequence of messages**: the system prompt, what you said earlier, what the model answered, and the results of tools. That sequence is the messages array; one continuous conversation is stored in a **session**.
@@ -190,29 +194,41 @@ Each line is the plain version first, the technical note second. The technical n
190
194
  *Technical: `baseCooldownMs: 300000`, doubling per consecutive failure up to `maxCooldownMs: 1800000`; the service's own `Retry-After` wins; parse failures and cancellations never cool an engine down.*
191
195
 
192
196
  9. **It can read the page, not just the snippet.**
193
- Search returns links; `web_fetch` returns text, and reroutes when a direct request fails.
194
- *Technical: a three-rung ladder — direct (5 s cap) → local tunnel → AnySearch extraction (needs a key); each page keeps at most 1 MiB (`1048576` bytes), with the rest truncated and flagged.*
197
+ Search returns links; `web_fetch` returns readable text, and reroutes when a direct request fails.
198
+ *Technical: a three-rung ladder — direct (5 s cap) → local tunnel → AnySearch extraction (needs a key); the raw body keeps at most 1 MiB (`1048576` bytes), which is then converted (item 10) and cut to `fetchMaxLength`.*
195
199
 
196
- 10. **It works on a restricted network.**
200
+ 10. **A fetched page comes back as readable Markdown, not raw HTML.**
201
+ Navigation, scripts, footers and forms are dropped; headings, paragraphs, lists, code blocks, quotes, links, images and tables stay — tables become Markdown tables. A page longer than the default 20,000 characters ends with one line naming the original length.
202
+ *Technical: `fetchFormat` (`markdown` by default, or `text`) and `fetchMaxLength` (default `20000`, range 1000–200000). Measured on `docs.python.org/3/library/urllib.parse.html`: 119,614 characters of raw HTML became 40,891 characters of body Markdown, and the tool returned 20,080 characters plus one truncation line. The body is chosen from `<article>`, then `<main>`/`role=main`, then the densest text container; relative links become absolute. **No resume**: the host's `WebFetchRequest` carries only `url` — no offset, no maxLength, no format — so a page cannot be continued from the middle; raise `fetchMaxLength` or fetch a narrower page.*
203
+
204
+ 11. **It works on a restricted network.**
197
205
  With a tunnel running and set as the system proxy, nothing needs configuring. With the tunnel off, it falls back to direct requests instead of failing.
198
206
  *Technical: proxy sources in order — the `proxy` setting, then `DSH_WEB_SEARCH_PROXY` / `HTTPS_PROXY` / `ALL_PROXY` / `HTTP_PROXY`, then macOS `scutil --proxy`; HTTP proxies only (`CONNECT`), not `socks5://`.*
199
207
 
200
- 11. **Every setting is editable in the UI.**
201
- Twenty controls in four groups; a save takes effect on the next search.
202
- *Technical: all twenty fields carry `.volatile()`, so the Host commits them in place instead of remounting the plugin.*
208
+ 12. **Every setting is editable in the UI.**
209
+ Twenty-four controls in five groups; a save takes effect on the next search, apart from the few the plugin reads once at mount.
210
+ *Technical: all twenty-four fields carry `.volatile()`, so the Host commits them in place instead of remounting the plugin; `cacheTtlMs` and `cacheEntries` are re-read per use, while the fetch provider takes `proxy`, `timeoutMs`, `fetchFormat` and `fetchMaxLength` at mount and the cache wiring takes `cachePersist` and `cacheFile` there too.*
203
211
 
204
- 12. **Failures are explained, not silently empty.**
212
+ 13. **Failures are explained, not silently empty.**
205
213
  When nothing succeeds, the error lists which engines were tried and why each one failed.
206
214
  *Technical: failures are classified `SEARCH_THROTTLED`, `SEARCH_BLOCKED`, `SEARCH_TIMEOUT`, `SEARCH_NETWORK`, `SEARCH_PARSE`, `SEARCH_ABORTED`; they are return values, not exceptions — only caller cancellation throws.*
207
215
 
208
- 13. **Packages, repositories, papers and Q&A are one hop away.**
216
+ 14. **Packages, repositories, papers and Q&A are one hop away.**
209
217
  For "is this npm package still maintained?" or "is there a paper on RAG?", the model can call `domain_search` against eight dedicated sources instead of hoping a web search finds them.
210
218
  *Technical: eight keyless sources — `github`, `npm`, `pypi`, `arxiv`, `wikipedia`, `hn` (Hacker News), `stackexchange`, `crates` (crates.io); arguments `query` (required), `sources` (optional array, default = every enabled source), `limit` (optional, default 5, capped at 20). Each source has its own deadline (8000 ms by default), one failure never affects the others, and the failed names come back in `failures`. Measured with `react` at limit 3: github 3, npm 3, pypi 1, arxiv 3, wikipedia 3, hn 3, stackexchange 3, crates 3. Known limits: PyPI has no public search API, so it matches the **exact package name** taken from the first word of the query; Wikipedia searches one language only (`zh` for CJK, otherwise `en`). A host without `ctx.tools` does not register the tool and the rest of the plugin works as usual.*
211
219
 
212
- 14. **You can look only at recently published results.**
220
+ 15. **You can look only at recently published results.**
213
221
  Choose "last 24 hours / 7 days / 30 days / 1 year" in Settings and only results inside that window stay — useful for news and releases.
214
222
  *Technical: `timeRange` takes `off` (default) / `day` / `week` / `month` / `year` — 24 hours, 7 days, 30 days, 1 year. Publication dates are parsed from each engine's results (Bing's snippet date and `news_dt`, 360's `发贴时间`, DuckDuckGo's timestamp and snippet, AnySearch's `date:` trailer). DuckDuckGo receives the window as `df=d/w/m/y`; Bing's and 360's parameters were measured to have no effect (the page shows the option selected while the results stay the same), so those two and AnySearch filter locally. With a window on, only **dated** results inside it are kept; only when not one dated result falls inside does the undated list come back, so the answer is never empty for that reason alone.*
215
223
 
224
+ 16. **Several questions, one call.**
225
+ For "compare these three libraries", the model can hand `search_many` one to five queries and get each query's results back separately; up to three run at once, and one failed query does not take the others down.
226
+ *Technical: `queries` (1–5, required), `limit` (default 5, 1–20), `withSites` (default false; true attaches each result's hostname). `SEARCH_MANY_CONCURRENCY = 3`. Each query goes through the same engine layer, cache and cooldown tracker as `web_search`; a thrown search becomes that query's `failures` entry, never the batch's. Use `search_many` for several questions at once, `web_search` for one, `domain_search` for packages, papers or Q&A. The tool's deadline is 30 s and it is concurrency-safe. It registers as `web_search_many` if the name is taken, and is not registered at all when the host has no `ctx.tools`.*
227
+
228
+ 17. **The cache survives a restart.**
229
+ Ask the same question again after restarting DSH within ten minutes and the previous result is still there; nothing is sent to the engines.
230
+ *Technical: `cachePersist` (default `true`) and `cacheFile` (default empty = `~/.dsh/cache/dsh-web-search/cache.json`). Only the summary — `url` / `title` / `snippet` / `publishedAt` — is written, never page bodies. 4 MiB default budget, LRU eviction, atomic write (sibling temporary file + `rename`); a corrupt, foreign or unreadable file loads as an empty cache and never stops the plugin. Two processes writing the same file is last-writer-wins, and a file is never left half-written. `cacheTtlMs` and `cacheEntries` apply immediately, no restart.*
231
+
216
232
  ## Installing and enabling it
217
233
 
218
234
  ### What you need
@@ -299,13 +315,13 @@ Optionally, add a config row:
299
315
 
300
316
  After changing code or profile config, **quit DSH entirely and open it again** (on macOS, quit from the menu or press `Cmd+Q`; closing the window is not enough). Modules are evaluated once per process, and a running process never sees a file you edited underneath it.
301
317
 
302
- Values saved from the Settings page do not need a restart; see the next section.
318
+ Values saved from the Settings page apply immediately; only `cachePersist` and `cacheFile` need the plugin to reload.
303
319
 
304
320
  ### Step 5: confirm it works
305
321
 
306
322
  Any one of these:
307
323
 
308
- 1. **Look at Settings**: Settings → "Web search (Bing)" should show four groups of controls. That means the plugin loaded and the Host accepted its schema.
324
+ 1. **Look at Settings**: Settings → "Web search (Bing)" should show five groups of controls, 24 in all. That means the plugin loaded and the Host accepted its schema.
309
325
  2. **Look at the log**: no warning such as `1 entry did not activate`.
310
326
  3. **Just try it**: ask something that obviously needs the network, such as "search for a DSH plugin tutorial and give me the links". Watch whether `web_search` is called and whether the links are real.
311
327
 
@@ -313,14 +329,15 @@ If the Settings card appears but search still fails, start with [docs/troublesho
313
329
 
314
330
  ## Using the Settings page
315
331
 
316
- Open Settings → "Web search (Bing)". Twenty controls, arranged in four groups. Each group owns one kind of decision:
332
+ Open Settings → "Web search (Bing)". Twenty-four controls, arranged in five groups. Each group owns one kind of decision:
317
333
 
318
334
  | Group | Controls | What it decides |
319
335
  | --- | --- | --- |
320
336
  | Engines and transport | 6 | Which engines run, how long to wait, whether to use a proxy, AnySearch credentials and vertical domain |
321
337
  | Results and ranking | 7 | How many results, how far back to look, how they merge and sort, how much official docs and fresh content count |
322
338
  | Dedicated search | 3 | Which sources `domain_search` queries, results per source, deadline per source |
323
- | Cache and cooldowns | 4 | How long results are remembered, how many, and how long a throttled engine rests |
339
+ | Fetching | 2 | What `web_fetch` returns — Markdown or plain text — and how many characters |
340
+ | Cache and cooldowns | 6 | How long results are remembered, how many, whether the cache is saved to disk, and how long a throttled engine rests |
324
341
 
325
342
  ### Engines and transport (6)
326
343
 
@@ -353,29 +370,41 @@ Open Settings → "Web search (Bing)". Twenty controls, arranged in four groups.
353
370
  | Results per source | `domainLimit` | `5` | How many results `domain_search` takes from each source, 1–20 |
354
371
  | Deadline per source | `domainTimeoutMs` | `8000` | Per-source deadline in ms, at least 500; raise it to 15000 on a slow network |
355
372
 
356
- ### Cache and cooldowns (4)
373
+ ### Fetching (2)
357
374
 
358
375
  | Control | Key | Default | Plain meaning |
359
376
  | --- | --- | --- | --- |
360
- | Cache lifetime | `cacheTtlMs` | `600000` | Ten minutes; `0` disables the cache |
361
- | Cache size | `cacheEntries` | `100` | How many queries to remember; the least recently used goes first |
377
+ | Output format | `fetchFormat` | `markdown` | `markdown` keeps headings, lists, tables, links and code; `text` keeps only the words. Leave it on `markdown` for reading; choose `text` for a smaller, plainer result. Applies immediately |
378
+ | Characters kept | `fetchMaxLength` | `20000` | How much converted text one fetch returns, 1000–200000; the cut is announced with the original length. Raise it to `50000` for long documentation. Applies immediately |
379
+
380
+ ### Cache and cooldowns (6)
381
+
382
+ | Control | Key | Default | Plain meaning |
383
+ | --- | --- | --- | --- |
384
+ | Cache lifetime | `cacheTtlMs` | `600000` | Ten minutes; `0` disables the cache. A change applies to the next search |
385
+ | Cache size | `cacheEntries` | `100` | How many queries to remember; the least recently used goes first. A change applies to the next search |
386
+ | Save the cache to disk | `cachePersist` | `true` | Keep the cache in a file so a restart can still hit it; turn it off to leave no trace on disk or when the disk is read-only. Applies immediately |
387
+ | Cache file | `cacheFile` | empty (default path) | Where the snapshot goes; empty means `~/.dsh/cache/dsh-web-search/cache.json`. Read when the plugin mounts |
362
388
  | Cooldown (ms) | `baseCooldownMs` | `300000` | First rest after throttling, five minutes |
363
389
  | Max cooldown (ms) | `maxCooldownMs` | `1800000` | Ceiling for the doubling backoff, thirty minutes |
364
390
 
365
391
  ### My advice
366
392
 
367
- **For normal use: leave all twenty alone. The defaults are the tuned values.**
393
+ **For normal use: leave all twenty-four alone. The defaults are the tuned values.**
368
394
 
369
- Only three situations are worth changing:
395
+ Only these situations are worth changing:
370
396
 
371
397
  - **Your proxy is not detected**: put the local proxy address in `proxy`, for example `http://127.0.0.1:7890`. HTTP proxies only — swap `socks5://` for the client's HTTP/mixed port.
372
398
  - **You want AnySearch's higher limits or page extraction**: fill in `anysearchApiKey`, and add `anysearchDomain` if you want one vertical index.
373
399
  - **You only want `domain_search` to hit a few sources**: narrow `domainSources` to a set such as `github,npm`; empty means all eight.
400
+ - **A fetched page is cut too early, or you want plain text**: raise `fetchMaxLength` to `50000`, or set `fetchFormat` to `text`.
401
+ - **You do not want a cache file on disk**: turn `cachePersist` off, or point `cacheFile` somewhere else.
374
402
 
375
- Two things to keep in mind:
403
+ Three things to keep in mind:
376
404
 
377
- - A save is immediate and applies to the next search, no restart.
378
- - **Ranking changes can be hidden by the cache**: the cache key holds `lead|wait|limit|engine set|time window|query`, not the ranking knobs. Ask the same question within ten minutes and you get the old order. Rephrase it, or wait ten minutes. Changing the window changes the key, so an old entry cannot interfere.
405
+ - A save is immediate and applies to the next search, no restart — except the controls the plugin reads at mount (`fetchFormat`, `fetchMaxLength`, `cachePersist`, `cacheFile`), which need a restart.
406
+ - **Ranking changes can be hidden by the cache**: the cache key holds `lead|wait|limit|engine set|time window|query`, not the ranking knobs. Ask the same question within ten minutes and you get the old order — and with persistence on, restarting DSH does not clear it. Rephrase it, or wait ten minutes. Changing the window changes the key, so an old entry cannot interfere.
407
+ - **A page cannot be continued from the middle**: `web_fetch` makes one request per URL, so if the tail is missing, raise `fetchMaxLength` or fetch a narrower page.
379
408
 
380
409
  ## FAQ
381
410
 
@@ -403,6 +432,16 @@ Four usual causes:
403
432
  3. **It needs a login or a subscription**: fetching carries no session of yours.
404
433
  4. **Direct requests time out**: 5 seconds direct, then the tunnel; if neither works and no key is configured you get an error, and with a key there is one more rung, server-side extraction. The message names each rung's failure.
405
434
 
435
+ ### Why does fetched content differ from what the browser shows?
436
+
437
+ Because `web_fetch` is not a browser:
438
+
439
+ - It keeps the **article**, not the page: navigation, ads, scripts, footers, forms and hidden nodes are dropped before conversion. Headings, paragraphs, lists, code, quotes, links, images and tables survive.
440
+ - It does **not run JavaScript**: a page that renders its body client-side comes back nearly empty, while the browser fills it in.
441
+ - It **cuts the tail**: converted text beyond `fetchMaxLength` (20000 characters by default) is replaced by one line naming the original length, so a long page stops early.
442
+
443
+ If the text stops too soon, raise `fetchMaxLength` (`50000` is a reasonable value). If what comes back is an empty shell, the page probably needs a browser — try its print view, a static mirror or the site's API instead.
444
+
406
445
  ### Why do the results differ from what I get searching by hand?
407
446
 
408
447
  Because it is not the same operation:
@@ -419,6 +458,14 @@ Because a window removes results whose date falls outside it. With a window on,
419
458
 
420
459
  The one exception is the fallback: if not a single **dated** result falls inside the window, the undated results come back so you are not left empty-handed. So a windowed search returns somewhere between zero and eight results, not a fixed eight.
421
460
 
461
+ ### Will the cache give me the same old answer forever?
462
+
463
+ No. Each entry expires `cacheTtlMs` after it was written — ten minutes by default — so a repeated question inside that window gets the old answer, and the next one goes back to the engines. Three things keep it bounded:
464
+
465
+ - `cacheTtlMs: 0` turns the cache off, and `cacheEntries` (default 100) caps how many answers are kept, evicting the least recently used;
466
+ - the disk snapshot stores **summaries only** (`url`, `title`, `snippet`, `publishedAt`), never page text;
467
+ - expiry is absolute: an entry reloaded from disk keeps its original deadline, and an expired one is dropped at load, so restarting DSH cannot extend an entry's life.
468
+
422
469
  ### `domain_search` or `web_search` — which one?
423
470
 
424
471
  Depends on whether you want a **page** or an **entry**:
@@ -428,11 +475,15 @@ Depends on whether you want a **page** or an **entry**:
428
475
 
429
476
  They chain: `domain_search` finds the package or paper entry, then `web_fetch` opens its homepage or README for the detail.
430
477
 
478
+ ### What is the fastest way to ask several questions?
479
+
480
+ Use `search_many`. One call carries one to five `queries`, runs three at a time, and returns each query's results separately; `limit` (default 5) caps results per query, and `withSites: true` adds each result's hostname. A failed query becomes its own `failures` line instead of ending the batch. Use `web_search` for a single question, and `domain_search` when the target is a package, a repository, a paper or a Q&A thread.
481
+
431
482
  ### I changed a setting, why did nothing happen?
432
483
 
433
484
  Three cases:
434
485
 
435
- - **Changed on the Settings page**: saved immediately and read on the next search. But if you ask the same question again within ten minutes you may hit the cache and see no difference — try rephrasing.
486
+ - **Changed on the Settings page**: saved immediately and read on the next search, except `fetchFormat`, `fetchMaxLength`, `cachePersist` and `cacheFile`, which are read when the plugin mounts. But if you ask the same question again within ten minutes you may hit the cache and see no difference — try rephrasing; with persistence on, a restart does not clear the cache either.
436
487
  - **Changed the profile's `cordis.patch.yml`**: requires a full restart of the harness.
437
488
  - **Changed `src/*.ts`**: requires `npm run build` and then a restart. Source edits without a build still run the old code in `lib/`.
438
489
 
@@ -442,7 +493,7 @@ A little, and it is usually worth it:
442
493
 
443
494
  - A measured search takes 1.4–3.8 s (English waits a bit longer, Chinese is faster); a cache hit is nearly instant.
444
495
  - The model is waiting for a tool result during that time, not hanging; if the deadline arrives, it gets whatever has already come in.
445
- - The real cost is context: search results and page text both consume the model's input, so longer pages make the turn more expensive and leave less room for everything else. That is why search returns eight results by default and a page is capped at 1 MiB.
496
+ - The real cost is context: search results and page text both consume the model's input, so longer pages make the turn more expensive and leave less room for everything else. That is why search returns eight results by default and a fetched page is converted and then cut at 20,000 characters (the raw body is capped at 1 MiB first).
446
497
 
447
498
  ### Why does the model sometimes not search at all?
448
499
 
@@ -461,13 +512,13 @@ web_search call
461
512
  → filter by the hosts allow-list; drop viaProxy engines when no proxy exists
462
513
  → skip engines in cooldown (but never end up with none)
463
514
  → decide the leading region from the query script (ASCII → global, CJK → cn)
464
- → look up the cache (key: lead|wait|limit|sorted engine ids|time window|normalised query)
515
+ → look up the cache in memory, then the on-disk snapshot (key: lead|wait|limit|sorted engine ids|time window|normalised query)
465
516
  → run the primary tier concurrently, release on the first satisfied condition
466
517
  → if nothing came back, run the fallback tier
467
518
  → each engine filters locally by timeRange (DuckDuckGo also gets df pushdown)
468
519
  → fuse per region, rank with RRF, slice to limit
469
520
  → filter once more after the merge (undated rows return only when nothing dated is inside)
470
- → store in the cache, return { sources, failures, proxy }
521
+ → store in the cache, mark the snapshot dirty, return { sources, failures, proxy }
471
522
  ```
472
523
 
473
524
  A tier releases on whichever comes first: every engine settled; every engine of the leading region settled (success or failure); enough sources from enough contributors; the soft deadline. Engines still running are aborted and reported as `still running at deadline` or `aborted once enough results had arrived`.
@@ -518,6 +569,17 @@ Details:
518
569
 
519
570
  **Cooldowns** exist because scraping public result pages is not a supported interface. `SEARCH_THROTTLED` (429), `SEARCH_BLOCKED` (403, other 4xx, a failed echoed-query check), `SEARCH_TIMEOUT` and `SEARCH_NETWORK` put an engine to rest. `SEARCH_PARSE` (upstream markup drift — retrying cannot help) and `SEARCH_ABORTED` (the caller cancelled) never do. The duration prefers the service's own `Retry-After` (delta seconds or an HTTP date, read from the final redirect hop); otherwise `baseCooldownMs × 2^(consecutive failures − 1)`, capped at `maxCooldownMs`. One success clears the streak. If every engine is cooling, the run still uses them, so the engine set is never empty.
520
571
 
572
+ ### The cache on disk
573
+
574
+ `cachePersist` (default `true`) turns the file on; `cacheFile` defaults to empty, which resolves to `~/.dsh/cache/dsh-web-search/cache.json` (a leading `~` expands). The document is `{ version: 1, savedAt, entries: [{ key, expiresAt, sources }] }`, and each entry carries only the summary — `url`, `title`, `snippet`, `publishedAt`; any other property a source picked up is dropped at the serialisation boundary, and page bodies never reach the file.
575
+
576
+ - **Write**: the snapshot goes to a sibling temporary file (`<file>.<pid>.<n>.tmp`) and is then `rename`d over the target, which is atomic on one filesystem; a failure before the rename leaves the previous document intact and removes the temporary file. Writes are throttled (2 s) and also flushed on dispose and on process exit.
577
+ - **Read**: an absent, unreadable, malformed or foreign-`version` file, and any individually invalid entry, all collapse to an empty cache; loading never throws and never stops the plugin from mounting.
578
+ - **Bound**: the serialised document is trimmed in LRU order until it fits 4 MiB (`DEFAULT_CACHE_FILE_MAX_BYTES`), and `cacheEntries` still caps the in-memory map.
579
+ - **Expiry**: `expiresAt` is absolute, so an entry reloaded from disk keeps its original deadline; expired entries are dropped at load, and a restart cannot extend a cache entry's life.
580
+ - **Two processes**: last writer wins. One process may replace another's newer snapshot with an older one, but a half-written file is impossible.
581
+ - **Live reading**: `cacheTtlMs` and `cacheEntries` are functions read on every use, so a Settings edit applies to the next search; `cachePersist` and `cacheFile` decide the wiring at mount, so changing those two needs a restart.
582
+
521
583
  ### The three-rung web_fetch ladder
522
584
 
523
585
  The `http-proxy` fetch provider tries three routes in order, each only after the previous one failed:
@@ -530,7 +592,18 @@ The `http-proxy` fetch provider tries three routes in order, each only after the
530
592
 
531
593
  The direct deadline is deliberately shorter than the tunnel's: a blocked host usually fails by timing out, and spending the whole budget on a route that cannot work leaves the route that can with nothing.
532
594
 
533
- Other details: only `http(s)` URLs are accepted; each page keeps at most `1 MiB` (`1048576` bytes), the rest truncated and flagged `truncated`; a non-2xx response is a **result**, not a failure, and comes back with its status code and body; only when all three rungs fail does the provider throw, and the message names each rung's reason.
595
+ Other details: only `http(s)` URLs are accepted; the raw body keeps at most `1 MiB` (`1048576` bytes) before conversion, the rest truncated and flagged `truncated`; a non-2xx response is a **result**, not a failure, and comes back with its status code and body; only when all three rungs fail does the provider throw, and the message names each rung's reason.
596
+
597
+ ### From HTML to Markdown
598
+
599
+ The conversion is dependency-free (`src/markdown.ts`): a tolerant tokenizer and tree builder, because real pages are not XML; a pruning pass that drops chrome tags, comments and hidden nodes; a body-selection pass that prefers `<article>`, then `<main>` / `role=main`, then the densest text container; and a renderer for headings, paragraphs, lists, code, quotes, rules, links, images and tables.
600
+
601
+ - `markdown` keeps the syntax; `text` keeps only the words. Entities are decoded and relative URLs become absolute.
602
+ - Tables become Markdown tables; a table outside the chosen article root is dropped with the rest of the furniture.
603
+ - The character budget applies **after** conversion: `truncateFetchText` cuts at `fetchMaxLength`, trims trailing whitespace, avoids splitting a surrogate pair, and appends `[truncated at 20000 chars; original length 40891 chars; the rest is not shown]`.
604
+ - Measured on `docs.python.org/3/library/urllib.parse.html`: 119,614 characters of HTML → 40,891 characters of Markdown → 20,080 characters returned.
605
+ - The provider is built with `fetchFormat` and `fetchMaxLength` at mount, so changing either on the Settings page needs a restart; search-side settings are re-read per call.
606
+ - **No resume**: the host's `WebFetchRequest` has exactly one field, `url` — no offset, no maxLength, no format. A fetch therefore always starts at the top; the only remedies are raising `fetchMaxLength` or fetching a narrower page.
534
607
 
535
608
  ### The domain_search pipeline
536
609
 
@@ -553,6 +626,20 @@ Two sources need a word of their own:
553
626
 
554
627
  Registration goes through the host's real tool registry, read with `ctx.get('tools')` rather than `inject` — an inject entry is a hard dependency, so a deployment without a tool service would fail to mount the plugin and `web_search` would disappear with it. When the registry is absent the tool is not registered and the rest of the plugin works as usual. If another plugin already owns the name, it registers as `web_domain_search`.
555
628
 
629
+ ### The search_many pipeline
630
+
631
+ ```
632
+ search_many call
633
+ → queries 1–5 (required), limit default 5 (valid 1–20), withSites default false
634
+ → mapLimit over the queries with SEARCH_MANY_CONCURRENCY = 3
635
+ → each query: the same engine layer, cache and cooldown tracker as web_search
636
+ → a thrown query becomes { sources: [], failures: [message] }; the batch continues
637
+ → results keep input order; withSites adds the lower-cased hostname as `site`
638
+ → return { results, truncated }
639
+ ```
640
+
641
+ Each query sees the configured `limit` as its ceiling: a caller may ask for fewer results, never for more. The tool's own deadline is 30 s, it declares itself concurrency-safe, and it registers as `web_search_many` when another plugin owns `search_many`. Like `domain_search`, it is read from `ctx.get('tools')`: a host without a registry does not register it, and the rest of the plugin keeps working.
642
+
556
643
  ### Proxy
557
644
 
558
645
  Node's global `fetch` reads neither the macOS system proxy nor `HTTPS_PROXY` — unless the process started with `NODE_USE_ENV_PROXY=1`, which cannot be switched on from inside a running process. So when a proxy is needed, the plugin speaks to it directly: send `CONNECT`, upgrade with TLS for `https:` targets, write the request by hand, and decode `Content-Length`, chunked, gzip, deflate and brotli, following up to four redirects. `307`/`308` keep the method and body; any other `3xx` downgrades a `POST` to a bodyless `GET`.
@@ -565,6 +652,10 @@ Resolution order: the `proxy` setting → the environment (`DSH_WEB_SEARCH_PROXY
565
652
  - **Google is unreachable from this network.** A tunnelled `www.google.com/search` returns a JavaScript shell (94 KB, 1–3 links); headless Chrome is redirected to `google.com.hk` with a reCAPTCHA. `gbv=1`, a consent cookie and a mobile user agent all failed. Eight public SearXNG instances were unusable too (429/403/not JSON).
566
653
  - **It depends on upstream HTML structure.** A redesign breaks a parser; that engine records a reason and yields to the next one rather than silently returning nothing.
567
654
  - **Titles, links and snippets only** — no page text. That is `web_fetch`'s job.
655
+ - **A fetched page cannot be resumed from an offset**: the host's `WebFetchRequest` carries only `url`, so text beyond `fetchMaxLength` needs a higher limit or a narrower page.
656
+ - **Extraction keeps only what looks like the article**: a page built entirely by JavaScript comes back nearly empty, and content outside the chosen article root — a sidebar table, say — is dropped with the navigation.
657
+ - **`search_many` is capped at five queries and three in flight.**
658
+ - **Two processes sharing one cache file write last-writer-wins**: neither can corrupt the file, but one process's newer snapshot can be replaced by another's older one.
568
659
  - **A time window can only filter results that carry a date**: dates are parsed from result pages, so an engine that prints none leaves that row without an age; when nothing dated falls inside the window the undated list comes back. Bing's and 360's time parameters were measured to have no effect; only DuckDuckGo's `df=` is a real pushdown.
569
660
  - **PyPI matches exact package names only**: it has no public search API, so the plugin uses the first word of the query; a typo yields an empty answer.
570
661
  - **Wikipedia searches one language only**: `zh` for CJK queries, `en` otherwise.
@@ -583,9 +674,16 @@ Resolution order: the `proxy` setting → the environment (`DSH_WEB_SEARCH_PROXY
583
674
  | Per-source deadline `domainTimeoutMs` | `8000` ms |
584
675
  | Cache TTL `cacheTtlMs` | `600000` ms = 600 s |
585
676
  | Cache entries `cacheEntries` | `100` |
677
+ | Cache persisted `cachePersist` | `true` |
678
+ | Cache file `cacheFile` | empty = `~/.dsh/cache/dsh-web-search/cache.json` |
679
+ | Persisted cache budget | `4194304` bytes = 4 MiB |
586
680
  | Per-engine timeout `timeoutMs` | `15000` ms |
587
681
  | `web_fetch` direct deadline | `5000` ms |
588
- | Bytes kept per page | `1048576` (1 MiB) |
682
+ | Raw bytes kept per page | `1048576` (1 MiB) |
683
+ | Fetch format `fetchFormat` | `markdown` (or `text`) |
684
+ | Characters kept per fetch `fetchMaxLength` | `20000` (1000–200000) |
685
+ | `search_many` queries / concurrency | 1–5 / 3 |
686
+ | `search_many` limit | `5` (1–20) |
589
687
  | RRF constant `rrfConstant` | `60` |
590
688
  | Base cooldown `baseCooldownMs` | `300000` ms = 5 min |
591
689
  | Max cooldown `maxCooldownMs` | `1800000` ms = 30 min |
@@ -593,14 +691,14 @@ Resolution order: the `proxy` setting → the environment (`DSH_WEB_SEARCH_PROXY
593
691
  ### Development and tests
594
692
 
595
693
  ```sh
596
- node --test test/*.test.ts # the whole suite (239 tests)
694
+ node --test test/*.test.ts # the whole suite (283 tests)
597
695
  node --test test/proxy.test.ts # one file
598
696
  npm run typecheck # tsc --noEmit
599
697
  npm run build # compile src/ to lib/
600
698
  npm run check # typecheck + tests
601
699
  ```
602
700
 
603
- The suite runs offline — 239 tests: parser tests replay captured fixtures from `test/fixtures/` (normal and empty responses for all eight dedicated sources), transport tests open no sockets, proxy tests cover `parseProxy`, `proxyFromEnv` and the pure response framer, and time-window tests pin the date with an injected `now`.
701
+ The suite runs offline — 283 tests: parser tests replay captured fixtures from `test/fixtures/` (normal and empty responses for all eight dedicated sources), transport tests open no sockets, proxy tests cover `parseProxy`, `proxyFromEnv` and the pure response framer, time-window tests pin the date with an injected `now`, and the newer suites cover Markdown extraction (`markdown.test.ts`), batch search (`search-many.test.ts`) and the disk snapshot (`cache-persist.test.ts`).
604
702
 
605
703
  CI (`.github/workflows/ci.yml`) runs `npm ci`, `npm run typecheck`, `npm test` and `npm run build` on Node 24 for Ubuntu and macOS, then checks that the build emitted `lib/index.js` and `lib/index.d.ts` and that both `apply` and `Config` import.
606
704