Phantom is an AI co-worker that operates its own computer, maintains persistent memory, uses MCP tools, and builds software and infrastructure in its workspace. It is for people who want an agent that can continue work across sessions through Slack or web chat while keeping its workspace separate from the user's computer. The catalogue skills and instruction support workflows for Phantom.
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 skills add ghostwright/phantom --skill ritualgit clone --depth 1 https://github.com/ghostwright/phantomWrote 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/ghostwright/phantom/ritual)<a href="https://agentmods.dev/skills/ghostwright/phantom/ritual"><img src="https://agentmods.dev/badge/skills/ghostwright/phantom/ritual/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/ghostwright/phantom/ritual"><img src="https://agentmods.dev/badge/skills/ghostwright/phantom/ritual.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00018 | $0.01224 |
| Opus 5.5 | $0.00007 | $0.00490 |
| Sonnet 5.5 | $0.00004 | $0.00245 |
| Haiku 4.5 | $0.00002 | $0.00122 |
Grade A, and why
ritual 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 1mo 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 — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ritual: latent patterns to scheduled jobs
Goal
Find recurring behaviors in the user's history that emerged naturally over time without being formalized as scheduled jobs, and propose turning them into first-class schedules. The user does not need to remember to do the thing; the agent does it for them and delivers the result where they are.
The test is "what does the user already do on a cadence that the agent could prepare for them so they do not have to start from scratch each time." Not "what should the user be doing". Only what they already do.
Steps
1. Pull the last 60 days of sessions
Call mcp__phantom-reflective__phantom_list_sessions with days_back: 60, limit: 200. Note the started_at timestamp, channel, and the first user message of each session if you can see it.
Success criteria: you have a list of 50+ sessions from the last two months with timestamps.
2. Look for temporal repetition
Cluster the sessions by:
- Day of week (Monday, Tuesday, ...).
- Time of day (bucket into morning, midday, afternoon, evening).
- Channel.
- Topic, if you can infer it from the first message.
A candidate ritual is a cluster where:
- Three or more sessions happened.
- They share day of week OR time of day (ideally both).
- They share topic or channel.
- They are spaced at roughly the same cadence (weekly, biweekly, monthly).
Example candidates:
- "Every Monday morning around 8:30 in #ops you ask for a standup."
- "Every second Friday in Slack DM you ask me to prepare a weekly review."
- "Every first of the month in #finance you ask for a cost breakdown."
Success criteria: you have identified 1-5 candidate rituals.
3. Verify with memory
For each candidate ritual, call mcp__phantom-reflective__phantom_memory_search with a query matching the topic and days_back: 60. Confirm that memory also shows the same pattern.
Discard any candidate that the session pattern suggests but memory does not support. Discard any where the cadence is off (the user did it three Mondays in a row, then stopped two weeks ago).
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.
- 1mo ago First seen · 102 lines · 18 tokens per session scan A 9d3d45d9a65b
ritual is a skill published in the GitHub repository ghostwright/phantom (1,483 stars, last pushed 4mo ago), licensed Apache-2.0. It adds 18 tokens to every session and 1,224 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
ui-control
Open a panel in agentglass's own window, read what is open, or read and change one of its settings. Use when the person asks you to "show me" a Settings page, the machine panel, the bench, a Git modal or a file in agentglass; when you need to know what they are looking at, what is in a chat or on the bench, or what a…
whiteboard-voice-notes
When the user shares a voice recording on the whiteboard, naturally recognize the auto-generated transcript and bring it up in conversation. If the transcript is missing/empty or the user explicitly asks for re-transcription, recover it with the whiteboardtranscribe tool.
scout-briefing
Launch a Scout morning briefing session as a background process. Auto-detects mode (morning briefing, consolidation, or weekend briefing) from the current time and day.
scout-consolidation
Launch a Scout consolidation session as a background process. Runs a lighter delta scan — captures what changed since the last run and updates action items and KB.
scout-dream
Launch a Scout dreaming session as a background process. Processes feedback, does KB deep work, and works on wishlist items.
ashlr-lean-tools
Highest-impact token-saving habits for Claude Code tool use. Read-once, batch edits, orient-before-search, skip verification reads, grep-before-read, concise output at high spend. Reduces session token cost 30–60%.