- Compute subagent cost from token breakdown instead of relying on
unpopulated proc.metrics.costUsd; extract actual model from subagent
messages for accurate pricing and mismatch detection
- Add aggregated "Subagents (combined)" row to cost table with arrow
navigation to the Subagents report section (no misleading breakdown)
- Filter <synthetic> model from token/cost tracking to eliminate zero rows
- Fix parseTestSummary to treat missing pass/fail count as 0 so clean
all-passing test runs are not dropped from trajectory analysis
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Skip indexing and lookup when tool call/result IDs are missing to
prevent collisions on the empty-string key in toolCallIndex.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- getPricing: use tokenized order-insensitive matching so Claude 3 era
models (e.g. "claude-3-opus-20240229") correctly match pricing key "opus-3"
- sessionAnalyzer: skip isSidechain messages in parentCost loop to prevent
double-counting with processSubagentCost
- CostSection: disable row expansion when token stats are missing
- tabSlice: guard against undefined sessionId/projectId in openSessionReport
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add AssessmentBadge component with hover tooltips explaining thresholds
- Add Key Takeaways summary section surfacing top actionable findings
- Add cost attribution stacked bar and per-token calculation breakdowns
- Add clickable navigation from takeaways and tool errors to detail sections
- Add theme-aware assessment colors via CSS variables for light/dark mode
- Collapse lower-priority sections by default for progressive disclosure
- Replace all hardcoded color hex values with CSS variable references
- Fix missing Fragment key in CostSection model table
- Add defensive division-by-zero guard in stacked bar calculation
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add interpretive assessment layer to session reports. Every metric section
now shows color-coded severity badges (green/amber/red) computed from
configurable thresholds, replacing raw numbers with instant signal.
Includes centralized reportAssessments utility, model mismatch detection,
switch pattern recognition, and 44 new tests.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add 7 missing sections and fix 4 incomplete ones that the Python
analyzer generates but the TypeScript port was missing:
New sections: skillsInvoked, bashCommands, lifecycleTasks,
userQuestions, outOfScopeFindings, agentTree, subagentsList
Fixed: compaction (was just count, now has summaryCount + note),
serviceTiers removed (not available in parsed types),
compactionCount replaced with compaction object
Adds InsightsSection UI component and 8 new tests.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Port all analysis logic from the Python script to TypeScript, running
entirely in the renderer process. Produces a typed SessionReport from
a SessionDetail with: token/cost analysis, friction signals, idle gaps,
conversation tree metrics, test progression, and thrashing detection.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add an export button to the TabBar header that lets users export
the current session as Markdown, JSON, or Plain Text. The button
appears between Search and Notifications, only for session tabs.
- sessionExporter.ts: formatters for all three formats + download trigger
- ExportDropdown.tsx: dropdown UI component with format selection
- TabBar.tsx: integration with conditional rendering for session tabs
- 51 new tests covering all formatters, edge cases, and download
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Updated the default notification triggers to set 'enabled' to false for the .env File Access Alert, Tool Result Error, and High Token Usage triggers, allowing for more controlled notification management.
Sessions without the cwd field (older JSONL format) were creating a
separate subproject group, even when all sessions with cwd shared the
same value. The orphan subproject got a relative fallback path that
failed git identity resolution, causing zero sessions to load on select.
Now only counts distinct real cwds when deciding whether to split,
treating cwd-less sessions as belonging to the same project.
- Updated ContextBadge to display the total count of tool outputs and task coordination items based on their breakdowns.
- Introduced FlatInjectionList component for a denested view of injections, allowing users to toggle between grouped and flat views in SessionContextPanel.
- Added state management for flat view toggle and integrated FlatInjectionList into the existing layout.
- Introduced new options to copy Session ID and Resume Command in the SessionContextMenu component.
- Added visual feedback for copied actions with appropriate icons and labels.
- Updated menu height to accommodate new items and maintain layout consistency.
- Updated `markAllNotificationsRead` and `clearNotifications` functions to support optional filtering by trigger name, allowing for more granular control over notification management.
- Adjusted the `NotificationsView` component to reflect the new scoped functionality, including updates to button labels and unread count displays based on active filters.
- Added tests to verify the behavior of scoped actions for marking notifications as read and clearing notifications.
* feat(package): add afterInstall script for chrome-sandbox permissions
- Updated package.json to include an afterInstall script that adjusts permissions for the chrome-sandbox on Linux.
- Added new afterInstall.sh script to ensure proper ownership and permissions for the sandbox file, enhancing security and functionality.
* chore(package): add deb configuration section in package.json
- Introduced a new "deb" section in package.json to specify post-installation scripts for Debian packages.
- Enhanced the package configuration for better support of Linux distributions.
- Updated package.json to include an afterInstall script that adjusts permissions for the chrome-sandbox on Linux.
- Added new afterInstall.sh script to ensure proper ownership and permissions for the sandbox file, enhancing security and functionality.
- Updated package.json to include an afterInstall script that adjusts permissions for the chrome-sandbox on Linux.
- Added new afterInstall.sh script to ensure proper ownership and permissions for the sandbox file, enhancing security and functionality.
User messages in subagent JSONLs lack the isMeta field, defaulting to false.
An unconditional `continue` in the !isMeta branch skipped tool result
collection for these messages, causing all subagent tools to show
"No result received". Now we check for tool_result blocks before continuing,
allowing them to fall through to the result collection logic.
- Included a Product Hunt badge in the README to promote claude-devtools and enhance visibility.
- Updated the README layout for better presentation and user engagement.
- Included a section for Docker deployment in the README, detailing how to run the application using Docker.
- Updated the instructions to enhance user accessibility for Docker users.
