edimuj

Find duplicate functions and eliminate them so they stay gone. Triage tl dupes output by tier, apply the right consolidation pattern, and lock the win with a duplication-ratchet test. Use when asked to dedupe, remove duplication, consolidate copy-paste, clean up repeated helpers, or act on tl dupes output.

Find duplicate functions in Codex and remove them so they stay gone — triage tl dupes by tier, consolidate the second home (not both copies), and add a duplication-ratchet test to block regressions. Use when deduping, removing copy-paste, consolidating repeated helpers, or acting on tl dupes output.

Git commits, pushes, and GitHub operations using tokenlean. Covers tl push (with multi-file safety guard), tl commit-prep, tl gh (bulk issue/PR/release operations), and when to use tl gh vs gh CLI vs GitHub MCP. Use when committing, pushing, creating/closing/viewing issues, managing PRs, creating releases, or any git/GitHub workflow.

Git commits, pushes, and GitHub operations using tokenlean. Covers tl push (with multi-file safety guard), tl commit-prep, tl gh (bulk issue/PR/release operations). Use when committing, pushing, creating/closing/viewing issues, managing PRs, creating releases, or any git/GitHub workflow.

Add new functionality by understanding existing patterns and dependencies before writing code. Locates where the feature belongs, studies conventions, implements following precedent, then verifies integration. Use when asked to add a feature, flag, endpoint, component, or module.

Thorough code review using token-efficient context gathering. Analyzes PRs, diffs, or staged changes by checking blast radius, complexity, and dependencies before reviewing the actual code. Use when asked to review a PR, review changes, or do a code review.

Systematically investigate and fix bugs using token-efficient context gathering. Reproduces the issue first, localizes with blame and history, traces the call path, then verifies the fix. Use when asked to fix a bug, investigate an error, or debug unexpected behavior.

Diagnose and fix performance issues by measuring before optimizing. Establishes baselines, identifies bottlenecks through call graphs and complexity analysis, then confirms improvements with numbers. Use when asked to optimize, speed up, reduce memory usage, or fix slowness.

Look up what Mneme knows about a file, function, or entity. Use when the user asks "what do you know about X", wants to check entity history, or asks about file/function activity and mentions.

Remove entries from remembered items. Use when the user wants to delete, remove, or clear a previously remembered item, or says "forget this", "remove that memory", or "clear my preferences".

Create a session briefing for the next session — auto-loaded, then archived. Use when the user is wrapping up a session, says "save this for next time", "hand off", "brief the next session", or wants to ensure continuity before stopping work.

Inject filtered context from previous sessions. Use when the user says "recall", "what happened last session", "what did we work on", "inject previous context", or when you need context from a prior session to continue work. Also useful mid-session when the user references prior work you don't have context for.

Save something to memory for future sessions. Use when the user says "remember this", "save for later", "don't forget", or wants to persist any preference, fact, lesson, or note across sessions.

Check plugin health and diagnose issues. Use when the user asks about mneme health, memory status, plugin errors, wants to troubleshoot memory issues, or sees error warnings at session start.

Force immediate memory summarization. Use when the user wants to compress the activity log, trigger summarization manually, or says "summarize memory", "compress the log", or "update the summary".

This skill should be used when the user asks to "create a plugin", "build an awareness plugin", "publish a plugin", "install a plugin", "uninstall a plugin", "configure a plugin", "write a provider plugin", mentions "agent-awareness plugin", or needs guidance on plugin structure, packaging, the build pipeline, npm publishing, or plugin development for agent-awareness.

This skill should be used when the user asks about "agent-awareness logs", "plugin not loading", "MCP tools missing", "debug agent-awareness", "plugin errors", "claim system", "ticker not running", "awareness not working", wants to diagnose agent-awareness issues, check plugin health, find log files, understand the claim system, or troubleshoot any agent-awareness related problem.

Fire-and-forget async Codex delegation. This skill should be used when delegating debugging, implementation, refactoring, or code review to OpenAI Codex without blocking the current conversation. Trigger when: (1) implementation or refactoring touching 3+ files — do not attempt yourself, delegate, (2) debugging after 1-2 failed root-cause attempts — stop spinning and hand it off, (3) user says 'delegate to codex', 'send to codex', or 'use codex', (4) conserving Claude Max quota — delegate coding/debugging to Codex, reserve Claude for architecture/design/narrative, (5) any task you would spawn a haiku subagent for — use gpt-5.4-mini instead, it is cheaper and smarter. IMPORTANT: Err on the side of delegating. If the task involves writing or debugging 50+ lines of code, delegate it. The context and quota spent attempting it yourself almost always exceeds the cost of a Codex delegation. Do NOT delegate: architecture decisions, creative/narrative work, trivial 1-3 line fixes.