* Update telemetry wrapper to 0.15.1

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* Update telemetry wrapper to 0.15.2

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

---------

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

)

Replace the uuid package with Node.js built-in crypto.randomUUID(). This removes the uuid dependency entirely, addressing the Dependabot alert for uuid v8 and avoiding compatibility issues with uuid v14+.

…icrosoft#1637) (microsoft#1641)

* fix: always prepend separator when appending noConfigScripts to PATH

EnvironmentVariableCollection.append() performs a literal string concatenation
and does not insert a path separator. The previous heuristic checked
process.env.PATH on the extension host and skipped the separator when that PATH
already ended with one. However, the PATH used by the integrated terminal can
differ from the extension host's PATH, so this check could incorrectly drop the
separator and glue the noConfigScripts directory onto the last entry of the
user's PATH (e.g. C:\Program Files\jreleaser\c:\Users\...\noConfigScripts).

Always prepend the separator. A trailing empty PATH entry (if the user's PATH
already ended with one) is harmless on both Windows and POSIX shells.

Fixes microsoft#1637

* test: add unit tests for noConfigScripts PATH append helper

Extract the PATH append-value computation into a vscode-free
src/pathUtil.ts module so it can be unit-tested in plain Node mocha.

Add test/pathUtil.test.ts covering:
- Correct separator on Windows (;), Linux (:), and macOS (:).
- The appended value always starts with a path separator (regression
  guard for microsoft#1637).
- The scripts directory remains a standalone PATH entry when the user's
  PATH last entry has no trailing separator (the exact scenario reported
  in microsoft#1637, e.g. C:\Program Files\jreleaser\\).
- A trailing separator on the user's PATH only produces a harmless empty
  entry, never glues another entry onto our scripts dir.
- The scripts directory is preserved verbatim at the end of the value.

All 9 tests pass under plain mocha; the file is also picked up by the
existing Electron-based test runner via the **/**.test.js glob.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* style: trim verbose comments in pathUtil and tests

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

---------

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Bumps [fast-uri](https://github.com/fastify/fast-uri) from 3.1.0 to 3.1.2.
- [Release notes](https://github.com/fastify/fast-uri/releases)
- [Commits](fastify/fast-uri@v3.1.0...v3.1.2)

---
updated-dependencies:
- dependency-name: fast-uri
  dependency-version: 3.1.2
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

…age model tools (microsoft#1643)

- Gate all 10 language model tools with `when` clauses so they only register
  when the Java Language Server is ready, and (for active-debug-session-only
  tools) when an active `java` debug session is in progress. This avoids
  loading failures and reduces the noise Copilot sees from this extension's
  tool catalog when the tools cannot actually run.

- Contribute a `chatInstructions` file (`javaDebugContext.instructions.md`)
  with a keyword-rich, on-demand description (no `applyTo`, to avoid burning
  context on every Java edit) that tells Copilot to activate the deferred
  Java debug tools via `tool_search_tool_regex` and routes the user request
  to one of the two skills below.

- Contribute two `chatSkills`, split by user-habit telemetry (launch/stop is
  ~52% of tool usage; inspection/step is ~28%):
  - `java-launch-troubleshooting` — start/stop a Java program and diagnose
    launch failures (mainClass missing, classpath, build errors, project
    not detected). Gated by `javaLSReady` so it is discoverable any time
    in a Java workspace.
  - `java-debug-inspection` — inspect variables, walk the stack, list
    threads, evaluate expressions, step in/over/out, continue, and manage
    breakpoints in an active Java debug session. Gated by
    `javaLSReady && inDebugMode && debugType == 'java'` so it only appears
    once a Java debug session is alive — keeping it reactive, not proactive.

* Harden Language Model Tool telemetry against PII leaks

Centralise all LMT telemetry through src/lmToolTelemetry.ts so user-supplied strings (target, expression, sessionName, file paths, class names, JVM stack traces, etc.) can no longer reach the telemetry pipeline. The new module exposes a typed sanitizedSend choke point that only accepts enums, booleans, numbers and opaque session IDs.

Telemetry changes:

- Drop sendError(error) on debug_java_application failure (stack trace leaked user class / method names).

- Strip PII fields from every existing event: target, sessionName, currentFile, currentLine, simpleClassName, detectedClassName, error: String(error), input.reason.

- Replace bare String(error) propagation with classifyError() -> ErrorCategory enum (mainClassMissing, classpathUnresolved, buildFailure, projectNotDetected, sessionAlreadyRunning, timeout, lsNotReady, noActiveSession, noSuspendedThread, noStackFrame, cancelled, other).

- Add per-invoke recording for all 10 tools with outcome, errorCategory, durationMs, and a tool-specific enum (targetType / breakpointKind / stepKind / scopeType / evalContext / removeScope). The previous build only emitted telemetry on the launch tool and the session-info tool.

- Add chatActivationSnapshot one-shot at registration time so we can measure adoption of the chat surfaces without per-turn cost (counts only).

- evaluate_debug_expression: the expression text is NEVER logged. Only the evalContext enum and outcome are emitted.

Policy:

- src/lmToolTelemetry.ts is now the only file in the LMT code path allowed to call sendInfo. The top-of-file policy comment is the single source of truth for what may be logged.

- The recorder is typed against ToolInvocationRecord so excess raw strings are rejected at compile time.

Validated with: npm run tslint, npm run compile.

* Address Copilot review on PR microsoft#1644

- classifyStep: unknown step operations now report 'unknown' instead of being silently mislabeled as 'over'. Also adds a runtime guard in debug_step_operation so an unknown operation no longer reaches commandMap[op]/executeCommand(undefined) or session.customRequest with an arbitrary string.

- recordToolInvocation: introduces a private normalizeToolInvocationRecord that keeps 'outcome' and 'errorCategory' in lock-step for the six shared terminal values (cancelled / timeout / lsNotReady / noActiveSession / noSuspendedThread / noStackFrame). Fixes the case where debug_java_application returns {success:false,message:'Operation cancelled by user'} but outcome was 'failure' while errorCategory was 'cancelled'.

- get_debug_stack_trace: empty-stack-frame early return now sets errorCategory='noStackFrame' alongside outcome (was only setting outcome).

- recordLaunchInternal: signature is now a discriminated union (LaunchInternalEvent) instead of (operationName: string, properties: Record<string, SafeValue>). Unknown event names and unexpected property keys are now rejected at compile time. Updated all 8 call sites.

- elapsedTime (string from .toFixed) split from elapsedMs (number) so the telemetry value is numeric and aggregable.

Bumps [@nevware21/ts-utils](https://github.com/nevware21/ts-utils) from 0.13.0 to 0.14.0.
- [Release notes](https://github.com/nevware21/ts-utils/releases)
- [Changelog](https://github.com/nevware21/ts-utils/blob/main/CHANGELOG.md)
- [Commits](nevware21/ts-utils@0.13.0...0.14.0)

---
updated-dependencies:
- dependency-name: "@nevware21/ts-utils"
  dependency-version: 0.14.0
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>