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 dcramer/agents --skill inbox-triagegit clone --depth 1 https://github.com/dcramer/agentsWrote 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/dcramer/agents/inbox-triage)<a href="https://agentmods.dev/skills/dcramer/agents/inbox-triage"><img src="https://agentmods.dev/badge/skills/dcramer/agents/inbox-triage/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/dcramer/agents/inbox-triage"><img src="https://agentmods.dev/badge/skills/dcramer/agents/inbox-triage.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 | $0.00058 | $0.02925 |
| Opus 5.5 | $0.00023 | $0.01170 |
| Sonnet 5.5 | $0.00012 | $0.00585 |
| Haiku 4.5 | $0.00006 | $0.00293 |
Grade A, and why
inbox-triage 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 today.
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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Inbox Triage
Apply the user's private policy through a reusable, auditable workflow. This skill owns mechanics and safety; category tastes, contact protections, and sender exceptions belong in private per-account policy. Use sender, actual To/Cc recipients, subject, content, and relationship history together. Leave genuinely ambiguous cases unchanged and ask for feedback.
Workflow
- Load private account rules. Verify the connected account. Read the supplied rules path, otherwise
$XDG_CONFIG_HOME/inbox-triage/accounts/<account-key>/rules.md, falling back to~/.config/inbox-triage/accounts/<account-key>/rules.md. The account key is a private local identifier. Use a legacyinbox-triage/rules.mdonly after confirming it belongs to this account; never combine accounts' preferences. Use references/rules-template.md for configuration, learning, and audit fields. If no policy exists, report "no personal policy loaded", inspect read-only, and ask for the missing choices; do not invent category defaults. Explicit current instructions can supply scoped policy for this run. Resolve private audit storage and the audit label before changing mail. - Pick the window. Use the requested window; for recurring runs, start at the previous scheduled run with about 30 minutes of overlap. Default: the last 24 hours. Deduplicate actions and reports using the private audit history; do not repeat successful unsubscribe requests.
- Search and inspect. Search inbox mail within the window, excluding confirmed "Skip in search" entries, and page through all results. Check timestamps against the exact window. Retrieve actual To/Cc headers and sufficient content and thread history before deciding; snippets, display names, a jobs recipient, and subject prefixes alone cannot establish solicitation or a relationship. Check relevant sent/history records when relationship context is missing. Exact-sender searches alone can miss relationships: also check relevant organization/domain history, including other accounts only when authorized for this task. A different person or address may represent an established relationship; record evidence privately with its account provenance without merging accounts' policies. If essential evidence is unavailable, leave the affected mail unchanged and report the inspection limit.
- Classify with the decision order below. Cleanup mutations apply only to messages currently in INBOX, never archived or sent mail included in a thread. History may be read for context only. Prefer message-level actions; use a thread-level action only after inspecting every affected message and confirming each is currently in the inbox and has the same permitted action. Leave mixed or uninspectable threads unchanged when message-level operations are unavailable. Explicit corrections to messages previously changed are the only exception to inbox-only mutation scope.
- Audit and act. Follow the mutation procedure below for spam, archive, unsubscribe, and corrections. Never label messages merely read or kept.
- Report and learn. Report confirmed outcomes and blocked operations in a private digest. Ask for feedback on genuinely ambiguous cases, leaving them unchanged until clarified; continue independent clear cases.
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- today First seen · 65 lines · 58 tokens per session scan A ea7ca595f248
inbox-triage is a skill published in the GitHub repository dcramer/agents (57 stars, last pushed today), licensed Apache-2.0. It adds 58 tokens to every session and 2,925 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 2026-10-16.
Other skills, from other repositories
orbit-notion
Open Orbit briefing skill — selected by the Orbit pipeline when Notion is the user's only connected connector, or when the user explicitly scopes their daily digest to Notion. Pulls the past 24 hours of document edits, comments, mentions, and database row changes from the user's authenticated Notion connection and…
agentmail
Use your assigned AgentMail inbox to read email tasks, explicitly send or reply, and check delivery. Provided automatically by your inbox assignment.
rowboat-spaces
How to work as a member of Rowboat Spaces, the team chat where people and agents work together. Load it once per session, before acting on any request that reaches you from Rowboat; it covers your setup, mentions and hand-offs, the tools, and files.
feishu
Work with Feishu or Lark bots, docs, sheets, bitables, approval flows, and OpenAPI/MCP setup without hardcoding credentials.
pinchtab-mcp
Use this skill when a task requires browser automation through PinchTab's MCP server connected to a remote browser instance. Covers navigation, element interaction, data extraction, form filling, multi-step flows, and session management via MCP tools.
Cortex
Operate Cortex, the LifeOS memory system — the typed Knowledge Archive (People, Companies, Ideas, Research with typed related: links) plus recall of prior work sessions, ISAs, and conversations. Search, add, harvest, develop, ingest, distill, graph-navigate, recall. USE WHEN cortex, knowledge, knowledge base, search…