agentleFS
Sign inStart free
maurice-jobstmaurice-jobst/ai-workbench/skills/session-open/SKILL.md

session-open

Skill0 starsChanged 47 days agoMITView on GitHub

Open a work session in a file-first AI workbench. Orients from git log, the desk file, scripts/check and the tracker, then briefs what matters now and proposes exactly one next step. Use at the start of every session in a repo whose AGENTS.md describes the session loop.

3 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions3 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 3

    1. vardan6/trackline★ 2same name
    2. Oldvictech/session-continuity★ 1same name
    3. vishnujchandran/.agents★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository6 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 3 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Run git rev-parse --show-toplevel and read AGENTS.md there.

    line 12
  2. Distinctive:

    Never substitute a default protocol for a missing one.

    line 19
  3. Distinctive:

    Read, in this order, and keep the output for the brief:

    line 23
  4. Very specific:

    Every command listed under ## Commands that is marked read-only and costs nothing.

    line 32
  5. Very specific:

    If git log --since="<most recent Monday> 00:00" --oneline is empty this is the week's first session: read the full desk file and the last …

    line 35
  6. Distinctive:

    check findings (a cap or a broken link first)

    line 44

“Common” is measured: kept by at least 6 in 10 of the 3 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

Session open

Orient, brief, propose one thing. Do not start work until the operator picks.

1. Preflight

Run git rev-parse --show-toplevel and read AGENTS.md there. Stop and name workbench-scaffold as the fix if any of these holds:

  • the folder is not a git repository
  • AGENTS.md is missing
  • AGENTS.md has no ## Working here section describing an open and a close

Never substitute a default protocol for a missing one.

2. Orient

Read, in this order, and keep the output for the brief:

  1. git log --oneline -n 20 --date=short, git status --porcelain and git branch --show-current.
  2. The desk file: STATUS.md, or the file AGENTS.md names instead.
  3. scripts/check (fall back to python3 scripts/workbench_lint.py when there is no check script). Findings go into the brief; do not fix them yet.
  4. The tracker, if AGENTS.md names one: open issues, newest first, and anything assigned or labelled blocked.
  5. Every command listed under ## Commands that is marked read-only and costs nothing. Name any you skipped and why.

If git log --since="<most recent Monday> 00:00" --oneline is empty this is the week's first session: read the full desk file and the last week of the decision log instead of skimming. Keep the explicit 00:00; a bare date anchors to the current time of day.

3. Brief

One short block, each category omitted when empty:

  1. check findings (a cap or a broken link first)
  2. uncommitted work found by git status, and the branch it sits on
  3. dates from the desk file that turned urgent
  4. contradictions between the tracker, the log and the desk file
  5. stalled waiting-on items

Headline what fits. On a fat day, review beats completeness: name what was parked.

4. Propose one step

End with exactly one proposal: the single piece of work this session should do, with the issue it belongs to and the branch name it will use. Prefer the desk file's first item under ## Now unless the brief found something more urgent, and say so when you deviate. Then wait. The operator picks; the session does not start work on its own.

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.

Check your setup, free