Fixes and refinements across issue tracking and memory analysis. • Fixed a crash that occurred when connecting a Jira or GitHub account. • The Create Issue form now reads each Jira project's own configuration: issue types and priorities come from the selected project, fields the project doesn't use are hidden, and required fields are flagged before you start writing. • Jira and GitHub now report the actual reason a request was rejected instead of a generic authentication error. • The memory utilization chart adds a p99 line, fills the available space, and no longer clips its labels or hover tooltip. • The main window keeps the size you gave it when opening and closing a profile, and the compact/expanded list setting is remembered between launches. • The configuration-change badge now keeps its before/after values, and the storage capacity banner clears as soon as you resolve it.

LogStitch 1.4 quiets down big AWS accounts. You can now hide any Lambda you never open — from the Function Navigator's context menu, or in bulk straight from a filtered search — so a 60-function account reads like a 6-function one. Hiding is only a view preference: hidden functions keep syncing and keep counting toward every analytics and cost rollup, they still turn up in global search with a badge, and every hide is undoable. MCP setup is rebuilt around the same idea of less fiddling. The settings tab is now MCP Server, with a client picker covering Claude Code, Claude Desktop, Cursor, VS Code, Codex CLI, Zed, LM Studio, Windsurf and more, each with its own setup steps, a copyable config snippet, and Reveal in Finder for the config file — worth having, since every client wants the same URL under a different key and several fail silently if you guess wrong. A new Test Connection button runs a real request against the server and tells you what actually failed when something is off. Elsewhere: Scan for New Profiles now has a proper home in the Profiles and + menus instead of an unlabelled toolbar button, searching a JSON field for a specific value dropped from roughly 47 seconds to under a tenth of a second on large databases, and LogStitch no longer animates while minimized or sitting behind another window, where it had been quietly burning CPU.

The Correlated Request Detail view is rebuilt around a flow map of the whole request, and the invocation list now navigates from the keyboard. THE STITCHED REQUEST VIEW • Follow a request across every function it touched on a flow map. Each hop shows duration with its cold-start share, retries, cost, and peak memory, under a plain-English verdict naming where it broke. Functions that were expected to run but never did appear as ghost hops, and inferred orderings are drawn dashed so they're never mistaken for proven causality. • The view now presents full-bleed over the invocation list and detail pane instead of in a fixed 800x600 dialog, so the flow map, stream, and line reader all get real width. • Click any line to open it in the inspector, pivot the whole chain on a JSON field from its context menu, or copy a Markdown summary or the visible log lines to the clipboard. The copied report names every active filter, so its scope is never overstated. KEYBOARD NAVIGATION • Move through invocations with the up and down arrow keys — each one loads as you land on it, no Return needed. Click a row or tab to the list to hand it focus; the arrows keep driving the function navigator otherwise. RELIABILITY • Chain filters on indexed fields no longer empty the stream, and a report generated with a function filter applied now gives the request's real start time rather than the first visible hop's. • Reopening a chain after switching invocations no longer re-presents the previous request. • Log lines can no longer collide on identity, which could drop a row from the stream or scroll to the wrong one.

Sharper pattern detection — stack traces from JSON-logging functions are now recognized, and pattern sets stay consistent for Lambdas that appear under more than one AWS profile. -Functions whose structured loggers emit the stack as a JSON array now have their stack traces extracted and grouped, instead of being reported as "no stack trace" despite carrying a valid trace -Running pattern detection on a Lambda that exists under multiple AWS profiles now resolves to one consistent function record, so you no longer see duplicate pattern sets split across profiles

Caught vs. uncaught errors — LogStitch now tells the difference between errors that crashed your function and errors it caught and logged. Your function list, Overview, and details views split the two apart, so a function that's "noisy but working" no longer looks like one that's down. What's new: • Errors are now split into uncaught failures and caught application errors, with the health dot keyed to what actually matters • New Error Class filter to narrow invocations to exactly the errors you're chasing • Throttled invocations now show as their own chip, separate from real errors • More reliable MCP connection that no longer stalls during long syncs • Fixes across Overview charts, log filtering, tooltips, and sidebar behavior

• Overview cards and function-list counts now load live from CloudWatch in ~500ms instead of waiting on local data. • Rows show live 7-day invocations, errors, and health at a glance. • Staleness chips flag when a chart's data is behind — tap to sync; selecting a function auto-syncs stale data. • MCP tools report live CloudWatch counts and freshness metadata.