Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/sebin-gg/zlog/zlognpx skills add sebin-gg/zlog --skill zloggit clone --depth 1 https://github.com/sebin-gg/zlogWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/sebin-gg/zlog/zlog)<a href="https://agentmods.dev/skills/sebin-gg/zlog/zlog"><img src="https://agentmods.dev/badge/skills/sebin-gg/zlog/zlog.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00150 | $0.01790 |
| Opus 5 | $0.00075 | $0.00895 |
| Sonnet 5 | $0.00030 | $0.00358 |
| Haiku 4.5 | $0.00015 | $0.00179 |
Grade B, and why
zlog scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
find -L ~ -maxdepth 4 \( -path "*/.gemini/*" -o -path "*/.cursor/*" -o -path "*/.claude/*" -o -path "*/.ollama/*" -o -path "*/.windsurf/*" -o -path "*/.codex/*" -o -path "*/.lm-studio/*" \) \( -name "*.log" -o -name "*.o How it starts
The opening of the file, as written. The whole thing — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Zlog Multi-Instance Storage Optimizer
Compress background tool logs across AI agents (~77%-99.9% space saved / fast zstd -15 fallback). Purge 0-byte empty logs. Keep transcript.jsonl intact. Safe for concurrent running agents. Skip tiny logs (<10KB) and text docs. Handle symlinks cleanly. Outputs single-line total summary.
Execution Intents
- Standard Cleanup: Triggered on "zlog", "compress logs", "clean up chat logs", "pack logs", or session wrap-up.
- Dry-Run Preview: Triggered on "preview zlog", "dry run", "test log cleanup".
- Deep Scan: Triggered on "find new AI agents", "scan disk for hidden AI logs".
Safety & Invariant Protection
- Process Locks: Query
fuser -s "$f"(Linux/WSL),lsof "$f"(macOS), or[System.IO.File]::Open(..., 'None')(Windows 11 PowerShell). Skip file if active process lock exists. - In-Flight Window: Skip files modified <60 seconds ago (
-mmin +1). - Protected Files: Exclude
*.jsonl,SKILL.md,README*,LICENSE*, and files <= 10KB.
Commands
Standard Compression & Empty Log Purge (Linux, macOS, WSL, Git Bash)
for d in ~/.gemini ~/.config/Cursor ~/.cursor ~/.ollama ~/.claude ~/.config/claude-code ~/.windsurf ~/.codex ~/.cache/lm-studio; do
if [ -d "$d" ]; then
find -L "$d" \( -name "*.log" -o -name "*.out" -o -name "*.trace" -o -name "*.txt" \) -empty -type f -delete 2>/dev/null || true
find -L "$d" \( -name "*.log" -o -name "*.out" -o -name "*.trace" -o -name "*.txt" \) -not -name "*.gz" -not -name "*.zst" -not -name "*.xz" -not -name "*.jsonl" -not -name "SKILL.md" -not -iname "README*" -not -iname "LICENSE*" -size +10k -mmin +1 -exec sh -c 'for f; do (command -v fuser >/dev/null 2>&1 && fuser -s "$f" 2>/dev/null) || (command -v lsof >/dev/null 2>&1 && lsof "$f" >/dev/null 2>&1) || (command -v zstd >/dev/null 2>&1 && zstd -15 -q --rm "$f") || (command -v xz >/dev/null 2>&1 && xz -9 "$f") || gzip -f "$f"; done' sh {} + 2>/dev/null || true
fi
done
dirs=(); for d in ~/.gemini ~/.config/Cursor ~/.cursor ~/.ollama ~/.claude ~/.config/claude-code ~/.windsurf ~/.codex ~/.cache/lm-studio; do [ -d "$d" ] && [ ! -L "$d" ] && dirs+=("$d"); done; [ ${#dirs[@]} -gt 0 ] && du -ch "${dirs[@]}" 2>/dev/null | tail -n 1 || true
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 67 lines · 150 tokens per session scan B 6b41c317efd4
zlog is a skill published in the GitHub repository sebin-gg/zlog (1 stars, last pushed 9d ago), licensed MIT. It adds 150 tokens to every session and 1,790 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
component-family-consistency
Buttons, inputs, pills, badges, calendars, and other interactive components form a visual family — they share the same border-radius, colour logic, shadow scale, border style, and spacing rhythm. Inconsistency between them breaks the sense of a coherent product. Use when building or reviewing a component library…
extract-design
Extract a complete design system — colors, typography, spacing, components, shadows, and W3C design tokens — from any live website using Dembrandt. Runs a headless browser against the URL and returns real computed values from the DOM. Use when you need a site's actual design tokens, want to reverse-engineer a visual…
information-architecture
In large applications, information architecture determines whether users can find, understand, and act on data. Naming matters. The UI should mirror the data model and signal how data can be transformed. Dangerous or irreversible changes always require a confirm dialog. Use when designing navigation, naming entities…
algorithmic-color-palette
Derive a full UI colour palette algorithmically from one or two brand colours. Darker and lighter variants for interactive states, desaturated greys from the brand hue for borders and backgrounds, and semantic colours that feel coherent with the brand rather than generic. Use when building a colour system from scratch…
brand-visual-language
A brand's visual tone — playful or serious, rounded or angular — should be consistent across all UI elements. Shape language in typography, border-radius, and iconography communicates personality before a single word is read. Use when establishing a design system, choosing icon libraries, setting border-radius tokens…
data-display-and-selection
Complex data deserves multiple view modes — grid, list, table — chosen by the user based on their task. Row and item selection should use large hit areas (the whole row or card, not just a checkbox). Selected state is communicated through a subtle background colour shift. Mass actions appear when items are selected.…