agentleFS
Runtime-3rr0rRuntime-3rr0r/canvas-scanner/skills/email-drafting/SKILL.md

email-drafting

Skill2 starsChanged 56 days agoMITView on GitHub

Use when drafting or editing the user's emails.

6 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

2 rules. 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 6 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Write for a human, human-sounding voice: natural, approachable, not stiff.

    line 19
  2. Distinctive:

    When he says "sent," open Outlook → Sent Items → the email, and DIFF his sent text against my draft — extend these lessons.

    line 25

“Common” is measured: kept by at least 6 in 10 of the 6 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Config: resolve <PLACEHOLDER> tokens from ~/.hermes/school-config.md (see canvas-course-sync → School config). Never edit SKILL.md with real values.

Email Drafting

Email drafting (<USER_NAME>'s style — learn from every sent draft)

There is NO reliable "type into Outlook compose" automation (the preview-pane compose field mapping is broken — typing got routed to To/Cc/body wrongly, twice). Always deliver emails as text in chat for copy-paste; reading Outlook (Sent folder) for diffing IS reliable.

<USER_NAME>'s style fingerprint (learned from his sent Chem-lecture-miss apology 9/9/26 — update as he edits more):

  • Apologize directly first, zero preamble ("I apologize for..." not "I'm writing to apologize for...")
  • Short declarative sentences, one idea each
  • Own the mistake fully — NO defensive hedging. But: do NOT be overly apologetic or personal; stay professional and brief (competent student, not begging)
  • NO em dashes — ever. They read as AI-generated. Use commas/periods instead.
  • Write for a human, human-sounding voice: natural, approachable, not stiff. Easier to build trust.
  • End forward-positive: next steps / readiness, not deficit ("so I don't fall behind and am ready for next week's content")
  • Close: "Best," + full name + student ID (<STUDENT_ID>)
  • Concise — every sentence earns its place
  • Professor email subject: COURSE-SECTION: topic (e.g. "<COURSE_CODE>-B: Missed September 9 Lecture")

When he says "sent," open Outlook → Sent Items → the email, and DIFF his sent text against my draft — extend these lessons.

Inbox catch-up ("catch me up on mail")

Reliable: open https://outlook.cloud.microsoft/mail/inbox, read the sidebar (desktop_preview read) to enumerate unread + previews. The MAIL READING PANE is flaky for sequential click-through: it re-pins to the first opened message and refs go stale (same compose/pane bug class).

Workflow: 1) read inbox list = headers + previews of everything unread; 2) report catch-up with priorities; 3) if user wants full body of any, they click it and I read the open pane (reading THAT is reliable). Don't burn turns retrying stale refs.

Archived mail context: keep transcripts of important emails in Classes/Announcements/-style files or notes when relevant.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.