my-show-info

name	my-show-info
description	Use when you need to check your Claude Code setup — quick lookup for installed skills, hooks, agents, session info, settings, and plugins. Also use for "show info", "what skills do I have", "list hooks", or "my setup".
argument-hint	< skill-name | hooks | agents | session | settings | plugins | help >
allowed-tools	Read, Glob, Grep, Bash(cat:*), Bash(ls:*), Bash(head:*), Bash(wc:*)

/my-show-info

Fast, low-token lookup for your Claude Code setup. Prints file contents or short summaries — avoids heavy LLM processing.

How It Works

Parse the argument. Match it against the lookup table below. Execute the matching action.

Default behavior: print raw file contents. Only summarize when the argument explicitly asks for it or when aggregating multiple files.

---

Lookup Table

Skills

Argument	Action
<skill-name> (e.g., my-prompt, my-git-sync)	Read ~/.claude/skills/<skill-name>/SKILL.md, extract and print only the ## Quick Help section. Do not read or process the rest of the file.
skills or all-skills	List all skill directories under ~/.claude/skills/ with each skill's description from frontmatter (one line per skill).

Hooks

Argument	Action
<hook-filename> (e.g., session-start, cost-tracker)	Read ~/.claude/hooks/<name>.sh and print it.
hooks or all-hooks	List all hooks from ~/.claude/hooks/ with the first comment line from each .sh file as its description.
stop-hooks	Read ~/.claude/settings.json, find all hooks under "Stop", print each hook's filename and read the first 5 lines of each script.
start-hooks	Same as above but for "SessionStart".
pre-tool-hooks or pretool-hooks	Same for "PreToolUse".
post-tool-hooks or posttool-hooks	Same for "PostToolUse".
compact-hooks or precompact-hooks	Same for "PreCompact".

Agents

Argument	Action
<agent-name> (e.g., skeptic, github-scout)	Read ~/.claude/agents/<name>.md and print it.
agents or all-agents	Read ~/.claude/agents/README.md and print it.

Session

Argument	Action
current-session or session	Find the most recent summary_*.md in the current project's .claude/sessions/ directory. Print it raw — no summarization.
sessions or all-sessions	List all session summary files for the current project with dates and file sizes.
memory or MEMORY.md	Find and print the MEMORY.md for the current project from ~/.claude/projects/<path>/memory/MEMORY.md.

Settings & Config

Argument	Action
settings	Print ~/.claude/settings.json.
plugins	Read ~/.claude/settings.json, extract the enabledPlugins object, print each plugin with enabled/disabled status.
env or environment	Read ~/.claude/settings.json, extract the env object, print each variable (mask values longer than 20 chars).
statusline	Print ~/.claude/statusline.sh.
permissions	Read ~/.claude/settings.json, extract permissions object, print allow/deny lists.

Project Context

Argument	Action
CLAUDE.md or claude-md	Print the current project's .claude/CLAUDE.md (or root CLAUDE.md).
global-claude-md	Print ~/.claude/CLAUDE.md.
ports or port-map	Print ~/.claude/projects/README.md.
deprecated	Print ~/code/github/claude-config/deprecated-skills.yaml.

Catch-All

Argument	Action
No argument or help	Print this lookup table as a quick reference.
Unrecognized argument	Fuzzy match: search skill names, hook names, agent names, and setting keys. If a close match is found, confirm with the user. If no match, say "not found" and suggest help.

---

Gotchas

• Plugin list comes from settings.json — disabled plugins still show (check enabled flag)
• Hook count may differ from settings.json if hooks were added by skills at runtime

Rules

• Minimal tokens: For current-session, memory, individual skill/hook/agent files — just Read and print. Do not summarize unless the user asks.
• One Read call per item: Don't chain multiple reads for a single lookup. If aggregating (e.g., all-skills), use parallel reads.
• No writes: This skill is read-only. Never modify any file.
• Fuzzy matching: If argument is prompt, match my-prompt. If scout, list github-scout. If session-summary, match the hook session-summary.sh.
• Current project detection: Use git rev-parse --show-toplevel to find project root for session/CLAUDE.md lookups.
• help argument convention: Every skill has a ## Quick Help section. When ANY skill is invoked with help as its argument (e.g., /my-prompt help, /bug-hunt help), read only the ## Quick Help section from that skill's SKILL.md and print it. Do NOT execute the skill's main logic. This convention applies to ALL skills — /my-show-info is just one way to access it.

Quick Help

What: Quick, read-only lookup for your Claude Code setup — skills, hooks, agents, session, settings, plugins. Usage:

• /my-show-info my-prompt — shows my-prompt's Quick Help
• /my-show-info stop-hooks — lists all Stop hooks with script previews
• /my-show-info current-session — prints latest session summary (raw, no LLM)
• /my-show-info agents — prints agents README
• /my-show-info plugins — lists plugins with enabled/disabled status
• /my-show-info settings — prints settings.json
• /my-show-info help — shows this reference Also works: /my-prompt help, /bug-hunt help — any skill with help shows its Quick Help section.