Catalyst Stats — Development skill for Claude Code
Show a compact read-only report of the current session — approximate token usage (exact when transcript usage is present), cache-read ratio, and recent session-degradation / failure-pattern alerts.
How to install Catalyst Stats
Installs to ~/.claude/skills/adrijshikhar-catalyst-catalyst-stats/SKILL.md
mkdir -p ~/.claude/skills/adrijshikhar-catalyst-catalyst-stats && curl -fsSL https://raw.githubusercontent.com/adrijshikhar/catalyst/HEAD/commands/catalyst-stats.md -o ~/.claude/skills/adrijshikhar-catalyst-catalyst-stats/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Catalyst Stats does
description: Show a compact read-only report of the current session — approximate token usage (exact when transcript usage is present), cache-read ratio, and recent session-degradation / failure-pattern alerts. No model call; numbers are computed by scripts/session-stats.sh.
Run `bash ${CLAUDE_PLUGIN_ROOT}/scripts/session-stats.sh "$CLAUDE_TRANSCRIPT_PATH"` (use the current session's transcript path; if unavailable, run with no argument). Print the rendered report verbatim to the user.
Alternatives in Development
- Pool Compare — Compare all Meteora DLMM pools for a token pair by APR, fee/TVL ratio, and volume 722 ★
- Memory Statistics — Show statistics about Claude's persistent memory file including total memories, storage size, and recent activ 332 ★
- 10 — Create a full-screen hero section with the following exact specifications 152 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Session Health
Manage the session-health skill — install/uninstall the UserPromptSubmit (per-turn) and Stop (session-end) hoo
Agent Stats
Show agent usage statistics — top agents, heatmap, outcome ratio
Transcriptmeter
What your agent sessions cost, from the transcripts on your disk: tokens, cache hit ratio, peak context, list-
Steve Name Brainstorming
Discover brand names from project sources and present only exact-match domain-available, collision-screened fi
Debug Helper
Use when something is broken and you need it diagnosed. A build failure, a runtime crash, a blank page, an unr
V Preferences
Manage Compound V decision preferences — your OWN dated past-reasoning memory (memory + challenge, never a clo
Related Agents
Infra Debugger
Use for diagnosing infrastructure and service issues — degradation, network problems, service failures, monito
Prod Incident Responder
Reads recent prod logs, picks the highest-impact recurring error, drafts a minimal fix PR with a regression te
Stats Analyzer
Analyzes ~/.claude/stats-cache.json and history.jsonl to produce a concise usage report with model token break