agentleFS
fizzfizz/claude-skills/morning-paper/SKILL.md

morning-paper

Skill0 starsChanged 8 months agoMITView on GitHub

Use at the start of a work session, when returning from being away, when the user says "what did I miss", "catch me up", "morning update", "afternoon paper", "saturday paper", "where are we", or when session-start completes and the user needs broader context beyond git state. Pulls Slack, Jira, Google Meet transcripts, and semantic search into a single briefing.

  • In this repositoryOne other file it gives its agents.

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 9 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    Distinctive · estimate

    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:

    Start of a work day or session (morning edition)

    line 24
  2. Distinctive:

    After session-start completes (session-start handles git state.

    line 28
  3. Distinctive:

    Read the team channel and any active DMs since last session.

    line 57
  4. Distinctive:

    Check the active sprint and any tickets assigned to you or recently updated.

    line 71
  5. Distinctive:

    Check for new transcripts since last session.

    line 97
  6. Distinctive:

    If new transcripts exist, rebuild the corpus and query for key topics:

    line 107
  7. Distinctive:

    After presenting, ask two things:

    line 169
  8. Distinctive:

    Not checking semantic search for context on active tickets

    line 181
  9. Distinctive:

    Not rebuilding the search corpus when new transcripts exist

    line 184

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “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

Morning Paper

Overview

A briefing that answers: what happened while you were away? Pulls from every source the user has — Slack, Jira, Google Meet transcripts, semantic search — and presents it as one readable summary. Like a newspaper: headlines first, details on demand.

The name is "morning paper" but the concept is edition-based — the time gap determines the depth:

EditionWhenDepth
Morning paperStart of work dayOvernight: Slack + Jira + Meet from yesterday
Afternoon paperBack from meetings/lunch/lifeLast few hours of Slack + Jira activity
Saturday paperMonday morning / back from days offFull multi-day digest, weighted by recency

Same pipeline, different time window. Scale the query depth to the gap.

When to Use

  • Start of a work day or session (morning edition)
  • Returning from meetings, lunch, errands (afternoon edition)
  • Monday morning or back from vacation (saturday/weekend edition)
  • User says "catch me up", "what did I miss", "where are we", "morning paper", "afternoon paper"
  • After session-start completes (session-start handles git state; this handles everything else)

When NOT to Use

  • Mid-session context check (just read the relevant source directly)
  • User only wants git state (that's session-start)

The Pipeline

digraph morning_paper {
  rankdir=TB;
  node [shape=box];

  slack [label="1. Slack\n(overnight messages)"];
  jira [label="2. Jira\n(board changes, new comments)"];
  meet [label="3. Google Meet\n(new transcripts)"];
  rag [label="4. Semantic Search\n(context recovery)"];
  brief [label="5. Assemble briefing"];
  present [label="6. Present to user"];

  slack -> jira -> meet -> rag -> brief -> present;
}

Step Details

1. Slack — what happened in the channels

Read the team channel and any active DMs since last session.

# Team channel — last 25 messages
slackcli conversations read #team-channel --limit 25

# Self-DM — check for drafts or notes left for yourself
slackcli conversations read #self-dm --limit 10

Extract: New messages from team members, decisions made, requests directed at you, blockers raised, anything that needs a response.

2. Jira — what moved on the board

Check the active sprint and any tickets assigned to you or recently updated.

# Your assigned tickets — current state
jira list -p PROJ -a "Your Name" -S "In Progress" 2>/dev/null
jira list -p PROJ -a "Your Name" -S "To Do" 2>/dev/null
jira list -p PROJ -a "Your Name" -S "Test" 2>/dev/null
jira list -p PROJ -a "Your Name" -S "Review" 2>/dev/null
jira list -p PROJ -a "Your Name" -S "Stalled" 2>/dev/null
jira list -p PROJ -a "Your Name" -S "Done" 2>/dev/null

# Recently updated tickets in the project (last 24h)
# This catches ALL status changes including "Won't Do" (which go-jira can't query by name due to the apostrophe)
jira list -p PROJ -q "updated >= -1d ORDER BY updated DESC" 2>/dev/null

For each ticket that changed since last session, get the latest comment:

lede TICKET

Extract: Status changes, new comments, new tickets created, blockers, anything you need to respond to or act on.

3. Google Meet — what was said in meetings

Check for new transcripts since last session. The DAG fetches them automatically.

# Check for recent transcripts
ls -lt ~/data/google_meet/dam_effective_date=$(date +%Y%m%d)/ 2>/dev/null

# If today has no transcripts, check yesterday
ls -lt ~/data/google_meet/dam_effective_date=$(date -v-1d +%Y%m%d)/ 2>/dev/null

If new transcripts exist, rebuild the corpus and query for key topics:

# Rebuild if new transcripts
make -C ~/semantic-search build

# Query for active work items
make -C ~/semantic-search query Q="action items decisions requests" LIMIT=10

Extract: Decisions made, action items assigned to you, architectural directions, requests from teammates.

4. Semantic search — context recovery

Query your semantic search tool for context relevant to today's planned work. Use ticket summaries and recent Slack topics as search terms.

# Query for each active ticket
make -C ~/semantic-search query Q="<topic from active ticket>" LIMIT=5

# Query for any names/topics that came up in Slack
make -C ~/semantic-search query Q="<topic from slack>" LIMIT=5

Extract: Prior decisions, context that informs today's work, things said in meetings that didn't make it to tickets.

5. Assemble the briefing

Structure like a newspaper — headlines first, then sections, most important at top.

## Morning Briefing — YYYY-MM-DD

### Headlines
- [1-line summary of most important thing]
- [1-line summary of second most important]
- [1-line summary of third]

### Slack Overnight
[Who said what, what needs a response, any decisions made]

### Jira Board
| Ticket | Summary | Status | What changed |
|--------|---------|--------|-------------|
[table of tickets that moved]

### Meeting Notes
[Key points from any new transcripts — decisions, action items, direction changes]

### Today's Plate
1. [Highest priority — why]
2. [Second priority — why]
3. [Third priority — why]

### Needs Response
- [anything that's waiting on you — Slack message, Jira comment, etc.]

6. Present to user

Output the briefing directly in the conversation.

After presenting, ask two things:

  1. "Want this in Marked 2?" — if yes, write the briefing to /tmp/<edition>-briefing-YYYY-MM-DD.md and open -a "Marked 2" /tmp/<edition>-briefing-YYYY-MM-DD.md. Edition is morning, afternoon, or saturday.
  2. "Anything you want to dig into before we start?"

Parallel Execution

Steps 1-3 are independent — run them in parallel using subagents or parallel tool calls when possible. Step 4 depends on what was found in 1-3 (search terms come from those results). Step 5 depends on all of 1-4.

Red Flags

  • Presenting raw Slack/Jira dumps instead of summarized briefing
  • Missing a source (forgot to check Slack, skipped Meet transcripts)
  • Not checking semantic search for context on active tickets
  • Burying the most important item below less important ones
  • Forgetting to include "Needs Response" items — these are time-sensitive
  • Not rebuilding the search corpus when new transcripts exist
  • Presenting the briefing as a wall of text instead of scannable sections

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.