AIRI is a self-hosted virtual AI companion that gives an AI character a voice, visual presence, memory, and the ability to interact with games and coding activity. People use it as a personal digital companion on the web, macOS, or Windows, including for voice chat and gameplay. The catalogue add-ons provide workflows for working with the AIRI project.
Borrowing it
Nothing to install: this file belongs to moeru-ai/airi. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/moeru-ai/airi/main/.agents/skills/analyze-io-traces/SKILL.mdgit clone --depth 1 https://github.com/moeru-ai/airiWrote 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/moeru-ai/airi/analyze-io-traces)<a href="https://agentmods.dev/skills/moeru-ai/airi/analyze-io-traces"><img src="https://agentmods.dev/badge/skills/moeru-ai/airi/analyze-io-traces/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/moeru-ai/airi/analyze-io-traces"><img src="https://agentmods.dev/badge/skills/moeru-ai/airi/analyze-io-traces.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00057 | $0.01073 |
| Opus 5.5 | $0.00023 | $0.00429 |
| Sonnet 5.5 | $0.00011 | $0.00215 |
| Haiku 4.5 | $0.00006 | $0.00107 |
Grade A, and why
analyze-io-traces scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Analyze IO traces
Tamagotchi saves each ended IO span as one JSON line when Record IO traces automatically is on. The setting is under Settings > System > Developer.
Find the files
Files are named <start time>.jsonl. A new file starts each time recording starts. Files older than seven days are deleted.
ls -t "$HOME/Library/Application Support/ai.moeru.airi/io-traces"/*.jsonl | head -3
ls -t "$HOME/Library/Application Support/@proj-airi/stage-tamagotchi/io-traces"/*.jsonl | head -3
The first path is the packaged app on macOS. The second is pnpm dev on macOS. If APP_USER_DATA_PATH is set, use $APP_USER_DATA_PATH/io-traces. If recording is off or nothing ran yet, the folder does not exist. Ask the user to turn it on and reproduce the problem.
Record shape
Each line is one span with traceId, spanId, parentSpanId, name, startTimeNano, endTimeNano, status.code (2 means error), attributes, and events. Time values are decimal strings in nanoseconds. Subsystems are in attributes["ai.moeru.airi.io.subsystem"].
A trace is one turn. A typed turn starts at Interaction turn. A spoken turn starts earlier, and one trace covers the whole path:
Wake word (kws) only when a wake word started the input
└── Voice input (voice-input) trigger: control, speech, or wake-word. outcome: committed, drafted, cancelled, or failed
├── Voice activity detection one span for each speech segment
├── Speech recognition (asr)
├── End of speech judgement one span for each pause that the chat model judged
├── Transcript rephrase
└── Interaction turn LLM inference, streaming control, speech turn, TTS synthesis, audio playback
A span ends before its parent can end, so a file can hold children whose parent is missing. The parent is still open, or the app stopped first. A Voice input with the outcome cancelled or drafted has no Interaction turn. Attribute ai.moeru.airi.io.cancelled marks work that the caller aborted, such as a judgement after the speaker continued.
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.
- 7d ago Changed · +23 lines ccd8ad6d54ba
- 18d ago First seen · 46 lines · 57 tokens per session scan A dcfef1072f08
analyze-io-traces is a skill published in the GitHub repository moeru-ai/airi (50,528 stars, last pushed today), licensed MIT. It adds 57 tokens to every session and 1,073 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-03.
Other skills, from other repositories
claude_code
This document describes how a Claude Code agent (or any agent using the Claude Code CLI pattern) invokes the Seiðr-Smiðja forge through the Rúnstafr CLI bridge.
debug-optimize-lcp
Guides debugging and optimizing Largest Contentful Paint (LCP) using Chrome DevTools MCP tools. Use this skill whenever the user asks about LCP performance, slow page loads, Core Web Vitals optimization, or wants to understand why their page's main content takes too long to appear. Also use when the user mentions…
systematic-debugging
Use when debugging a failing test, build error, or runtime issue that isn't immediately obvious. Guides a 4-phase root cause analysis instead of random fix attempts.
log-error-digest
Analyze log files to troubleshoot errors, identify peak error periods, and produce error clustering, frequency statistics, and time distribution reports. Supports JSON, syslog, and Nginx formats with automatic detection. Use when a user uploads a .log file and asks to analyze errors, find patterns, debug issues, or…
deck-retro
Run a fully local agent-deck retrospective over the user's own transcripts, Recall index and logs. Use when a user wants to find recurring failures or take their own finding from investigation through synthetic reproduction, a user-filed issue, a test-first fix and contributor PR handoff. Also use for weekly usage…
incident-response-incident-response
Coordinate production incident triage, mitigation, investigation, recovery, and follow-up. Use when responding to outages or other incidents that require structured operational coordination.