Skill Claude CodeCodex
Use when asked to analyze codescout tool usage, check for error patterns, audit tool health, generate a usage report, spot anti-patterns, or clear usage statistics to start fresh.
Instructions file
Instructions for mareurs/codescout, covering codescout, development commands, bug tracking, session intelligence trackers and querying active trackers (librarian).
Agent
Agent "claude-code" from mareurs/codescout, covering claude code, one-time setup, 1. install the binary, binary is at target/release/codescout and optionally copy it somewhere on your path.
Agent
Agent "copilot" from mareurs/codescout, covering github copilot, one-time setup, register codescout as an mcp server, enable agent skills and add workflow skills.
Agent
Agent "cursor" from mareurs/codescout, covering cursor, one-time setup, register codescout as an mcp server, add workflow skills and add the code-reviewer agent.
Agent
Agent "overview" from mareurs/codescout, covering agent integrations, feature comparison and guides.
Plugin Claude Code
Plugin marketplace listing 5 plugins: sdd, codescout-companion, claude-statusline, buddy, session-bridge.
Instructions file CodexOpenCode
AGENTS.md instructions for mareurs/claude-plugins, covering claude plugins marketplace, structure, config dir resolution, this machine (verified 2026-08-05) and active development focus.
Instructions file
Claude Code instructions for mareurs/claude-plugins, covering claude plugins marketplace, structure, config dir resolution, this machine — linux workstation (verified 2026-08-26) and active development focus.
Plugin Claude Code
A Himalayan-aesthetic bodhisattva companion for Claude Code, with 12 specialist masters on demand.
Command
You are acting as the user's primary bodhisattva. Follow these steps exactly.
Command
Consolidate accumulated memories — merge near-duplicates, archive stale entries, summarize tag-clusters, surface contradictions for resolution. Runs as a four-phase pipeline (rules shortlist → specialist judgment → user dry-run gate → apply). Pass a target (specialist alias, common, all) or one of the sub-commands…
Command
Create a new buddy specialist (global or project scope; builtin is frozen). Brainstorms with the user, drafts from the canonical skill template, previews what /buddy:summon will load (template + memory protocol + gates), then writes. Discovery is path-scan — no registration step needed. The argument is a freeform hint…
Command
You are releasing one or all summoned specialists. The argument passed by the user is $1.
Command
You are handling a /buddy:focus request. The argument is $1.
Command
You are installing the buddy statusline into the user's Claude Code settings.
Command
Run mid-session introspection on one or all currently summoned specialists. Unlike /buddy:dismiss, the specialist stays active afterward. Pass a specialist name to introspect just that one (e.g. /buddy:introspect hamsa); pass nothing to introspect all active.
Command
You are printing the buddy plugin reference card.
Command
Ask the currently active specialist(s) to save a memory about the given lesson. Pass the lesson as the argument — e.g. /buddy:remember in this repo, integration tests must hit a real database. The specialist decides global vs project scope and the slug.
Command
You are diagnosing the buddy plugin. Do the following steps.
Command
Summon a specialist bodhisattva to help with a specific craft. Describe who you need in plain language — e.g. "debug", "testing", "ML training", "architecture", "security", "refactor", "performance", "docs", "data leakage classic", "data leakage llm", "planning", "prompt". Some specialists have lenses; pass them as …
Command
You are uninstalling the buddy statusline integration from the user's Claude Code settings.
Hook
Runs before the agent uses a tool, running node. From mareurs/claude-plugins.
Hook
Runs when a session starts, running node. From mareurs/claude-plugins.