agentleFS
Edzard9Edzard9/stash/skills/study/lab-report/SKILL.md

lab-report

Skill0 starsChanged 8 days agoMITView on GitHub

Write a practicum or lab report in a fixed, proven house format: identity header, per-point observations each followed by a one-line "what it means", a short conclusion, the observation table, and Q&A-style evaluation answers. Use when the user asks for a lab report, practicum report, observation write-up, or says "buatkan laporan praktikum", "laporan praktikum", "hasil pengamatan", "tulis laporannya". Do NOT use for essays, journals, or research papers.

5 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

6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    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:

    If the user supplies an example report of their own, match its sections exactly and drop any section the example does not have.

    line 46
  2. Distinctive:

    Never invent an observation: Write only what the user actually did and saw: their screenshots, pasted outputs, and stated results.

    line 52
  3. Distinctive:

    Check against the rubric as text: Before delivering, compare the draft line by line with the task criteria.

    line 68
  4. Distinctive:

    Run the rubric check: every criterion named in the task appears in the draft, every number matches its source, every evaluation question …

    line 85
  5. Distinctive:

    Not an essay or journal writer.

    line 100
  6. Distinctive:

    Not a way to produce results the user did not get.

    line 102

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Lab Report

A practicum report is a form, not an essay. The lecturer checks it against a rubric, point by point, so the report wins by being complete, exact, and short, in the same shape every time. This skill is that shape, plus the rules that keep it honest. It comes from real submitted reports, and every rule below exists because breaking it once caused a real correction.

The format (always this skeleton)

Nama  : <full name>
NPM   : <student ID>
Kelas : <class>

LAPORAN PRAKTIKUM <number>: <topic>

Hasil Pengamatan
1. <what was observed, 1-3 relaxed sentences>
   Artinya: <what it means, 1-2 sentences>
2. ...

Kesimpulan
<one short paragraph, no new facts>

Lembar Hasil Pengamatan
| No | Kegiatan | Hasil |
|----|----------|-------|
| 1  | ...      | ...   |

Soal Evaluasi
1. <question>
   Jawab: <answer, direct, no padding>
  • The identity block uses the user's own name, student ID, and class. Take them from what the agent already knows about the user; if they are unknown, ask once and reuse them for every later report.
  • If the user supplies an example report of their own, match its sections exactly and drop any section the example does not have. Their accepted example outranks this skeleton.

Ground rules (never bend)

  • Never invent an observation. Write only what the user actually did and saw: their screenshots, pasted outputs, and stated results. Numbers are copied exactly from the source (0% packet loss stays 0%, an IP address is copied digit for digit), never rounded, improved, or filled in from what "usually happens". Missing data is a question to the user, not a gap to fabricate over.
  • Brevity is the format. Each observation point is a few sentences, each "Artinya" line is one or two. A draft that reads like a textbook chapter is wrong even if every fact in it is right; the user will send it back to be cut. When in doubt, cut.
  • The lecturer's file is the authority, with one exception. Modules and assignment files sometimes contain a part that is deliberately wrong. Follow the file as given; if a step is demonstrably wrong, propose the correct alternative and confirm with the user before deviating. Never silently fix it, never silently follow an error either.
  • Check against the rubric as text. Before delivering, compare the draft line by line with the task criteria. A step the user performed but the draft never states still counts as a gap until it is written in. The report is judged on what it says, not on what was done.
  • Hand back a finished file. Deliver the report as one simple DOCX (or the format the user asked for), already filled in, not as instructions for filling it in. Change nothing outside what was asked: no font, layout, or formatting "improvements".

Workflow

  1. Gather the material: practicum topic and number, the steps taken, the observed results (screenshots or pasted output), and the evaluation questions if the task has them. Ask for what is missing, once, in one message.
  2. Draft into the skeleton above, one observation point per step or result, in the order the work happened.
  3. Run the rubric check: every criterion named in the task appears in the draft, every number matches its source, every evaluation question has an answer.
  4. Deliver the finished file. A one-line summary of what is in it is enough; do not paste the whole report into the chat unless asked.

Language

The report body is written in relaxed academic Indonesian (Bahasa Indonesia) by default, because that is the format's origin; switch languages only when the user asks. Keep the technical terms in their usual form (command names, protocols, and addresses stay as-is).

What this skill is not

  • Not an essay or journal writer; the skeleton above is the whole product.
  • Not a way to produce results the user did not get. If the practicum failed, the report says what actually happened; a failed ping, honestly reported with its cause, is a complete observation.

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.