Always-on memory (structured user/feedback blocks, PROFILE.md, MEMORY.md) is injected into every system prompt but only ever grew, inflating per-turn context over time. This adds deterministic size control across all always-on sources:
- Injection budget: cap the always-on structured block by total chars and per-type entry count, keeping the most-recently-updated entries (LRU by Updated date) and disclosing how many were omitted
- Nightly consolidation: a dedicated scheduled pass merges duplicate/stale user & feedback entries via the LLM, preserving each entry's original Updated date; runs per owner bucket (shared + personal) with a per-run cap and a never-grow safety guard
- File ceilings: deterministic backstop truncates PROFILE.md / MEMORY.md at a section boundary when a rewrite overruns its budget
- Manual trigger endpoint for the consolidation maintenance task
All knobs under mate.memory.*; covered by unit tests.
- goal: continue (not skip) on max-iterations and evidence-insufficient turns.
A max-iterations turn grants a fresh iteration budget ("hard continuation"),
bounded per run and sized into the graph recursion ceiling, so a task too big
for one budget keeps going instead of stalling until the next user message.
- plan-execute: re-plan the remaining work on a step exception, and on a
signature-based stall (repeated failures / identical results / no usable
result) instead of advancing dependent steps with junk; bounded by a per-run
re-plan cap, with a graduated change-strategy nudge before the hard stop.
- plan-execute: auto-derive a goal from a genuine multi-step plan, seeding the
acceptance criteria from the plan steps, so the goal subsystem engages without
the model calling setGoal; broadcast goal_created so the UI hydrates.
- react: refund the iteration for setup-only rounds (load_skill / enable_tool)
so a tight budget is not eaten by the load-then-use two-step.
- ui: re-fetch the active goal when a turn finishes so a goal created or mutated
mid-conversation surfaces without depending on an SSE event.
- streaming: make retry backoff / total-time budget instance fields with a
test-only seam; clarify that the wall-clock budget (not max-retries) bounds a
sustained SERVER_ERROR loop to ~8 attempts, fixing the slow/flaky retry test.
The knowledge-base trust verification recorded wiki citations from every
non-readFile tool response by sniffing its JSON for a top-level title /
pages / chunks field. Tools like getGoalStatus return a top-level title,
which falsely populated the citation set and then forced [n] citations on
the final answer (otherwise flagged EVIDENCE_INSUFFICIENT). Gate citation
mining on the wiki_* tool name instead.
Likewise, the grounded answer contract (cite-or-refuse) was appended to
every ReasoningNode call unconditionally, degrading general agents that
have no knowledge base. Append it only when the agent has a wiki_* tool
bound, scoping the strict regime to KB-grounded scenarios.
Adds a regression test asserting a non-wiki tool with a top-level title
creates no wiki citations.
- rescan: keep the KB id as a string end to end so the 19-digit snowflake id
isn't truncated past Number.MAX_SAFE_INTEGER (rescan no longer 404s)
- lint: resolve [[...]] targets against page slugs AND titles like the viewer,
so a title reference to an existing page is no longer reported broken
- ingest: derive the slug deterministically from the title (no inconsistent
romanization), auto-recompute broken links once a KB finishes importing,
and reconcile dangling [[concept]] links — redirect to the covering page via
declared aliases, or demote to plain text when uncovered
- add the page aliases column migration for h2 / mysql / kingbase
- add useStreamingMarkdown: cap mid-stream markdown re-render to ~140ms,
full-fidelity render once the segment completes
- skip code-block language auto-detection while streaming (escaped plain
text), restore full highlighting on the final render
- defer echarts/mermaid blocks to a lightweight loading placeholder while
streaming so their parsers never run on truncated source
- bypass the render cache for streaming-mode output
- wire into ContentSegment and MessageBubble (content + thinking)
Surface the connected database product as a subtle chip in the Dashboard
header. SystemHealthService now reports a database label on /system/health
(reused by the front-end — no extra request), derived from a new
DatabaseBootstrapRunner.getDatabaseLabel() that reads the JDBC product name
once and normalizes it to a canonical label (MySQL / MariaDB / PostgreSQL /
H2, and 人大金仓 for the KingbaseES family), collapsing driver version noise.
Add ErrorClassificationTest regression cases for the AI-gateway retry
hardening: 5xx-before-4xx ordering (a proxy 502 whose body says
"bad request" stays retryable), Chinese / numeric provider billing
patterns, and DNS / TLS infrastructure-fatal detection.
Fix the cert-trust pattern while adding its test: Java's ValidatorException
emits "PKIX path building failed" with an uppercase PKIX and the error
chain is not lower-cased, so the previous lowercase pattern never matched
— an untrusted/expired cert chain fell through to the retryable
SERVER_ERROR bucket and was retried in vain instead of failing over.
The blanket SMALLINT->BOOLEAN flag-column conversion over-reached: six wiki
columns map to Integer (1/0) entity fields, not Boolean. On vanilla PostgreSQL,
reading a BOOLEAN into a JDBC int throws 'Bad value for type int : f', breaking
every wiki KB list / SSE chat. Revert only those six back to SMALLINT (V133/V134/
V135/V136/V146 in the PostgreSQL-family tree) with guard comments; genuine
Boolean-entity columns stay BOOLEAN.
The disk fallback's catch block logged at debug, so a failed scan
silently dropped recovered files — reproducing the same 'bot can't see
the file' symptom the fallback was added to fix. Promote to warn with
the full stack trace, matching cacheRecentFile's logging.
openAiCompatibleClientBuilder() was creating a new java.net.http.HttpClient
per request. Each instance spawns a selector thread and connection pool that
are never closed, exhausting the OS thread limit under frequent model-test
calls (e.g. DeepSeek provider).
Elevate the HttpClient to a static singleton so all OpenAI-compatible
provider requests share one connection pool and one selector thread.
Closesmatevip/mateclaw#328
loadRecentFilesFromDisk now filters out files older than RECENT_FILE_TTL_MINUTES
(60 min) so the disk fallback matches the Caffeine cache TTL and does not inject
stale attachments into future conversations.
Testability refactoring:
- recentFileCache: private → package-private (tests can seed the cache directly)
- chatUploadsRoot: new package-private Path field (tests redirect to @TempDir)
- loadRecentFilesFromDisk: add (Path dir, long cutoffMs) package-private overload;
private (String) wrapper delegates to it
- injectRecentFiles: private → package-private
New test class FeishuRecentFileCacheTest (14 cases):
- loadRecentFilesFromDisk: non-existent dir, empty dir, fresh files sorted
newest-first, stale files excluded by TTL, mixed fresh+stale, >5 files capped,
timestamp-prefix stripping, MIME guessing from extension
- injectRecentFiles: Caffeine cache hit, cache-miss disk fallback, empty disk,
duplicate-path dedup, image vs file part typing, null textContent guard
Relates to #325
The per-chat recent file cache (Caffeine, 60 min TTL) is purely
in-memory. After a process restart, GC eviction, or TTL expiry the
cache is empty, but the staged copies under data/chat-uploads/ survive
on disk. A follow-up text message that should have seen the cached file
instead found nothing — the bot replied as if no file was ever sent.
Changes:
- injectRecentFiles(): fall back to scanning data/chat-uploads/{id}/
when the Caffeine cache misses, sorted by last-modified time, capped
at RECENT_FILE_MAX_PER_CHAT (5).
- cacheRecentFile(): promote catch log from debug → warn with full
stack trace so silent download failures are visible in production
logs. Add entry-level info log for correlation.
- New helper loadRecentFilesFromDisk() + guessContentType().
Closes#325
Relates to #201
The KingbaseES change removed usingDbTime() unconditionally, which made
every deployment (MySQL/H2/PostgreSQL) fall back to app-server time for
distributed lock timing — reintroducing node clock-drift risk in
multi-instance setups. Re-enable usingDbTime() for databases in ShedLock's
built-in dialect map and skip it only for KingbaseES, which is not covered
and would otherwise throw at lock acquisition.
The KingbaseES JDBC driver is not on Maven Central; declaring it as a
required runtime dependency broke the default build for anyone without
the proprietary jar. Move it into an opt-in `kingbase` Maven profile
(build with `mvn package -Pkingbase`). No Java code imports the driver
classes — it is loaded at runtime via driver-class-name only, so the
default build no longer needs it.
Also drop `mateclaw.browser.ssrf-check-enabled: false` from the default
application.yml: the code default is true, and disabling the SSRF guard
globally is unrelated to KingbaseES support.
Add KingbaseES support as an opt-in profile: dedicated migration tree, bilingual seed data, runtime DbType detection (KINGBASE_ES / POSTGRE_SQL), and JDBC URL handling in the datasource manager.
The raw-materials surface was renamed to "Sources" when upload, paste,
directory scan and per-KB auto-sync were unified into one tab. The read-only
viewers' reading-toggle segment still carried the old "Raw materials" label,
so managers saw "Sources" while read-only viewers saw "Raw materials" for the
same panel. Point the segment at the same i18n key for a consistent name.
* feat(wiki): unify raw materials & source watcher into a Sources tab with per-KB auto-sync
The raw-material directory scan and the Advanced "source watcher" sub-tab were
the same engine (same kb.sourceDirectory, same WikiDirectoryScanService) split
across two surfaces with two editable directory inputs. Merge them into one
"Sources" tab (upload / paste / directory manual scan + auto-sync toggle +
the raw-material list) and drop the watcher sub-tab from Advanced.
Auto-sync is now per-KB opt-in: a new watcher_enabled column (V146) gates the
periodic scan per knowledge base. The server-global mate.wiki.watcher-enabled
stays as an ops master switch — a KB is auto-scanned only when both are on
(AND). Manual scans are unaffected. Scan interval stays global for now
(tracked separately).
Closesmatevip/mateclaw#314
* docs(wiki): document source-watcher global switch env vars
Expose MATE_WIKI_WATCHER_ENABLED / MATE_WIKI_WATCHER_INTERVAL_MS as
explicit placeholders in application-mysql.yml, .env.example and
docker-compose.yml, mirroring MATE_WIKI_ALLOWED_SOURCE_ROOTS. Notes the
AND semantics (global ops gate + per-KB toggle) so operators know the
global switch alone is not sufficient.
The KB workspace split into a reading view (pages + graph) and a manage view
(raw materials, config, transformations, advanced, recent-activity snapshot)
gated behind manage:wiki. That moved the raw-materials and recent-activity
surfaces — which read-only viewers (view:wiki without manage:wiki) could
previously browse — entirely behind the management gate, silently dropping
their access.
Re-surface both in the reading-view segmented toggle for viewers who lack a
management view. Managers keep the focused pages/graph toggle and still reach
these surfaces through the management view, so nothing is duplicated for them.
The content panels already render by activeTab, so this only widens the
reading toggle and the activeTab/readingTab types.
- Fail closed to a global fallback sandbox root when a conversation has no
per-workspace base path, instead of leaving file/shell tools unconstrained
- Refuse shell commands that delete the workspace root directory itself
- Block workspace-boundary escapes at the policy layer before the approval
prompt, not only at execution time
- Approval bar now shows the actual command / target path being approved
The pageType profile could only be edited as a raw JSON string. Add a
structured form editor (default) that builds the profile without writing JSON:
profile-level settings (fallbackType / allowAdditionalFields), an ordered page
type list (add via a small wizard, remove, reorder), and a per-type form for
label / description / layer / field schema, with stage prompts (route/create/
merge) and the markdown template folded into an "advanced" section that carries
inline descriptions and examples. A form/JSON toggle keeps the JSON view as the
final review surface; serialization preserves unknown keys for forward-compat,
and save/validate/reset reuse the existing endpoints (no backend change).
Closesmatevip/mateclaw#310
The KB workspace previously stacked all seven surfaces in one tab strip.
Split them by intent: a gear on each library card opens the management view
(raw materials, config, transformations, advanced, hot cache), while clicking
the card body opens the reading view (pages + graph). The reading view drives
page/graph via a header segmented control with the page tree shown only for
the page viewer; the two views share loaded data and toggle without refetch.
Closesmatevip/mateclaw#308
The V85 seed shipped a dev/test placeholder (sse + http://localhost:8085/sse
+ "Bearer ${CKJIA_MCP_KEY}") that can never connect, so the 参考价 price-
comparison skill was unusable until an admin hand-edited the row.
Add V144 (h2 + mysql) to rewrite the seed to the real CKJIA SaaS endpoint:
streamable_http + https://m.ckjia.com/api/ai/mcp, no auth header, connect/read
timeouts raised to 60s. Conditional on the row still carrying the dev
placeholder URL, so an admin who already pointed it at a private deployment is
left untouched; idempotent and leaves `enabled` opt-in.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Closes#289 — after an MCP server (re)connects, chat queries kept replying
"from memory" instead of calling MCP tools.
Root cause: agents snapshot their tool set at build time and are cached in
AgentService.agentInstances, but MCP server lifecycle changes never
invalidated that cache (unlike model-config / tool-guard changes which do).
A stale, tool-less agent graph survived until process restart.
Changes:
- Add McpServerChangedEvent; McpServerService publishes it on connect /
disconnect / reconnect / delete / (re)connect-failure / batch refresh /
startup init. AgentService listens and calls refreshAllAgents(), so the
next turn rebuilds against the live MCP tool set. Also closes the boot
race where the web server accepts requests before the @Order(200) MCP
init runner finishes.
- Make create/update/toggle connect asynchronously on a dedicated pool
("mcp-connect") so a slow/unreachable server can no longer freeze the
admin request; status returns immediately as "connecting".
- UI: render the new "connecting" status (pulsing amber dot), show a
friendly "connecting in background" toast, and poll until the status
settles (window widened to ~40s to outlast the default connect timeout).
- UI: MCP config modal no longer closes on outside/backdrop click — only
the × and Cancel buttons close it, so an accidental click can't discard
unsaved config.
Verified E2E: ckjia-shopping (参考价) MCP server connected at runtime with
no backend restart; the cached 通用助手 agent immediately enabled and called
ckjia_shopping_recommend, returning real product cards.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The webchat conversationId (webchat:<key8>:<visitorId>[:<sessionId>]) and the
derived username (webchat:<visitorId>) are written to VARCHAR(64) columns, but
visitorId had no validation and sessionId allows 64 chars — so a long visitorId,
or a legitimate 64-char sessionId, overflows the column and the getOrCreateConversation
INSERT throws (500 on /stream). Validate visitorId (charset + blank->UUID) and
fold the variable part into a stable hash when the derived id/username would
exceed 64 chars, keeping short ids byte-identical (backward compatible). Also
make listSessions filter on exact owner username, not just the conversationId
prefix, so system-owned rows can never leak via a crafted visitorId. Adds
boundary regression tests.
