A tool that creates a shareable year-in-review page about sensitive information exposed to an AI assistant. It presents exposure statistics, categories, risk, and related patterns in a single HTML file.

It is for generating 30-day, 90-day, or all-time recaps, including a demo with fictional data, and saving the result as a self-contained HTML page.

It turns scattered exposure history into a recap that is easier to inspect or share.

wrapped — Skill for Claude Code by sonomoshq

Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code

/plugin marketplace add sonomoshq/Canary

Claude Code

/plugin install canary

Made for: Claude Code.

Or install canary, the plugin that ships this one along with the rest of its 5 skills, 1 agent, 3 hooks.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/sonomoshq/canary/wrapped.svg)](https://agentmods.dev/skills/sonomoshq/canary/wrapped)

Your own site

<a href="https://agentmods.dev/skills/sonomoshq/canary/wrapped"><img src="https://agentmods.dev/badge/skills/sonomoshq/canary/wrapped.svg" alt="Measured on agentmods" height="20"></a>

Per session 37 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 625 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00037	$0.00625
Opus 5	$0.00018	$0.00313
Sonnet 5	$0.00007	$0.00125
Haiku 4.5	$0.00004	$0.00063

Measured 6d ago against content hash 73d1eb04ef30, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

wrapped 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 6d 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.

canary/skills/wrapped/SKILL.md · 56 lines

How it starts

The opening of the file, as written. The whole thing — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Canary Wrapped

Generate the self-contained, shareable "Wrapped" recap of the sensitive data the user has exposed to Claude — a Spotify-Wrapped-style HTML page with a big number, a risk grade, top leaked categories, the worst single day, the longest clean streak, and a persona reveal.

Parse $ARGUMENTS for a period and/or demo:

demo (in any position) → add --demo so the page renders ~140 realistic, fully-fake detections instead of real data. Use this when the user has little/no real history yet and wants to see what Wrapped looks like, or explicitly asks for a demo/sample.
30d (default), 90d, or all → passed as --period.

Run:

python3 "${CLAUDE_PLUGIN_ROOT}/scripts/wrapped.py" --period <30d|90d|all, default 30d> [--demo] --no-open

(--no-open keeps this non-interactive; mention the path so the user can open it themselves, and drop --no-open if the user explicitly asks to launch it in a browser.)

The script prints the output path on success — always relay that exact path back to the user. Tell them:

Where the file is (the printed path — normally ${CLAUDE_PLUGIN_DATA:-~/.sonomos}/canary-wrapped.html).
That it's a single self-contained HTML file, generated locally, with zero network requests — nothing was uploaded anywhere.
That it scrolls through several full-screen "scenes" (cover, the number, top categories, biggest spike, cleanest streak, persona, and an install CTA), and each scene is screenshot-ready on its own for sharing — screenshot just the one that's most interesting to them (the persona reveal is usually the shareable payoff).
If the period had zero detections, it still renders a clean, celebratory "0 exposures" version — that's expected, not a bug.

Detection Architecture

Wrapped reads the same leaks.jsonl detection log as the dashboard (/canary:leaked) and scores it against the same shared taxonomy (taxonomy.json), so the grade and persona shown here always agree with the dashboard's numbers for the same time window.

Read the full file on GitHub · 56 lines

Changes

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.

6d ago First seen · 56 lines · 37 tokens per session scan A 73d1eb04ef30

Subscribe to this mod's changes

wrapped is a skill published in the GitHub repository sonomoshq/Canary (2 stars, last pushed 6d ago), licensed MIT. It adds 37 tokens to every session and 625 once invoked, about $0.0002 per session on Opus 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 2026-08-31.

AI Safety Analytics and metrics Claude Code Security ai-privacy anthropic claude claude-code-plugin

second-thought

Use when an idea, decision, design, plan, response, tool call, or action is about to be accepted or executed, especially during brainstorming, convergence, and pre-execution where shallow agreement or goal drift may occur.

gg-mo/second-thought-agent-skill · 49 tokens

using-second-thought

Use when starting any conversation and before major decisions to ensure second-thought critique is auto-applied before commitment, especially in brainstorming, convergence, and pre-execution phases.

gg-mo/second-thought-agent-skill · 39 tokens

docs-drift

Documentation-drift audit for this plugin monorepo. Audit either each scoped plugin's unreleased changelog claims or its latest shipped release from the previous reachable plugin-name--v tag, then verify that plugin READMEs, docs, CLAUDE.md or AGENTS.md files, and root documentation tell the truth for the matching…

gtapps/claude-code-hermit · 170 tokens

cco-tools

Show what tools actually cost in tokens — learned per-tool averages from observed results, replacing the hardcoded MCP/Agent guesses.

egorfedorov/claude-context-optimizer · 27 tokens

test-run

Run plugin test suites in this monorepo and report a concise pass/fail summary. Optional plugin slug arg; without arg, runs all plugins under plugins/.

gtapps/claude-code-hermit · 35 tokens

github-pr-creation

Creates GitHub Pull Requests with automated validation and task tracking. Use when user wants to create PR, open pull request, submit for review, or check if ready for PR. Analyzes commits, validates task completion, generates Conventional Commits title and description, suggests labels. NOTE - for merging existing…

fvadicamo/dev-agent-skills · 75 tokens