agentleFS
917Dhj917Dhj/dkitchen/skills/goodmorning/SKILL.md

goodmorning

Skill10 starsChanged 8 days agoMITView on GitHub

When the user is back, close the night orders and report what needs their decision, how the orders were used, where work stalled, and what moved. Run in the orchestrator session.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If no orders were posted, say so in one line and report the same way, counting the night from the user's last message.

    line 17
  2. Distinctive:

    Use the user's language and this order.

    line 30
  3. Very specific:

    Improve pass: when the last pass in docs/agents/improve-kitchen-history.md is more than 14 days old, or there has been none, one line …

    line 37

≈ “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

Goodmorning

The user is back. Their attention is the scarcest thing in the kitchen, so the report leads with what only they can do. Progress comes last. The night is over, so the night orders end now: from this moment, approvals come from the user directly again.

Words follow the repo's vocabulary in .dkitchen.toml and docs/agents/.

1. Close the orders

Find the latest night orders on the orchestrator log ticket, with ding night show or by searching the ticket. Close them before anything else, with ding night close or one comment: <orchestrator signature>: night orders closed at <time UTC>. Roles that check the orders from now on find them closed.

If no orders were posted, say so in one line and report the same way, counting the night from the user's last message.

2. Gather

Gather the following from the sources, for the time since the orders opened:

  • the orchestrator log entries: each --item and the order number it relied on, and each --pending;
  • tickets with new comments, especially those now waiting on the user or blocked;
  • ding board and each job's state;
  • the hours each card sat idle, and why: what it was waiting for. Use job start and end times from logs or run records. Mark any figure you estimated.

3. Report

Use the user's language and this order. Leave out a section that is empty.

  1. Decisions for you: each a question with a link and your recommendation, so the user can answer in a word.
  2. Stalls: what stopped, when, why, and the idle card-hours it cost. A stall that happened despite the orders is the most useful line in the report, because it shows a gap in tonight's goodnight trace.
  3. Orders used: for each numbered order, used (where) or unused. Flag any action that went beyond an order, with its log line.
  4. Progress: one line per ticket that moved, with a link. Results go to the designer's reading where one exists; link it rather than interpreting.
  5. Make standing?: orders that were used and would read the same every night. An example is "the query-vector service may start on any free card". Propose each as a standing rule. On the user's yes, it goes into the repo's rules as they direct, and future goodnights stop asking it.
  6. Improve pass: when the last pass in docs/agents/improve-kitchen-history.md is more than 14 days old, or there has been none, one line suggesting /improve-kitchen. This report counts as the orchestrator's first report of the day.

4. After the report

Record the user's answers as user decisions on their tickets, the repo's usual way, and relay them. Then restart your watch.

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.