Stats
Actions
Tags
From flux
Captures learnings from the current session into the brain vault by invoking the flux-reflect skill.
How this command is triggered — by the user, by Claude, or both
Slash command
/flux:reflectflux/The summary Claude sees in its command listing — used to decide when to auto-load this command
## Step 0: Version Check (silent, non-blocking) Run the version check script silently. If an update is available, show a brief notice but continue: If `update_available` is true, print once at the start: Then continue with the command. Do NOT block or prompt - just inform. --- # IMPORTANT: This command MUST invoke the skill `flux-reflect` The ONLY purpose of this command is to call the `flux-reflect` skill. You MUST use that skill now. **User request:** $ARGUMENTS Pass the user request to the skill. The skill handles all reflection and brain-writing logic.
Run the version check script silently. If an update is available, show a brief notice but continue:
UPDATE_INFO=$(bash "${DROID_PLUGIN_ROOT:-${CLAUDE_PLUGIN_ROOT:-$(git rev-parse --show-toplevel 2>/dev/null || pwd)}}/scripts/version-check.sh" 2>/dev/null || echo '{"update_available":false}')
If update_available is true, print once at the start:
Flux update available (vLOCAL -> vREMOTE). Update Flux from the same source you installed it from, then restart your agent session.
Then continue with the command. Do NOT block or prompt - just inform.
flux-reflectThe ONLY purpose of this command is to call the flux-reflect skill. You MUST use that skill now.
User request: $ARGUMENTS
Pass the user request to the skill. The skill handles all reflection and brain-writing logic.
npx claudepluginhub nairon-ai/flux --plugin flux/reflectAnalyzes and validates task adherence, session progress, and completion using Serena MCP reflection tools. Provides recommendations and updates session metadata. Supports --type and analysis flags.
/reflectScans session history for corrections, extracts learnings, and updates CLAUDE.md files across project, global, and subdirectory tiers with human review. Supports dry-run, dedupe, organize, and scan flags.
/reflectAnalyzes filtered diary entries for patterns in user preferences, code decisions, solutions, and challenges, then synthesizes insights and proposes CLAUDE.md updates.
/reflectAnalyzes chat history, Claude instructions, commands, and config to suggest interactive improvements and implements approved changes, producing updated instructions.
/reflectGuides reflection on recent work like bug fixes or features to extract root causes, patterns, decisions. Records learnings via tk learn and tk decide.
/reflectAnalyzes conversation history and session ledger to suggest specific, actionable improvements to AGENTS.md based on patterns and feedback.