agentleFS
eklavyagoyaleklavyagoyal/wingman/skills/tracker/SKILL.md

tracker

Skill0 starsChanged 2 months agoMITView on GitHub

Show the application pipeline, check its integrity, and surface what needs action today

7 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

5 rules. 0 are common, 2 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 2

    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:

    Read tracker.md and list DATA_DIR/jobs/.

    line 15
  2. Distinctive:

    Every job folder has a row, every row points at a folder that exists

    line 17
  3. Distinctive:

    No duplicate rows for the same employer plus role

    line 18
  4. Very specific:

    No row with Applied set but status still Bewertet

    line 19
  5. Very specific:

    No row past its Next date without being surfaced

    line 21

“Common” is measured: kept by at least 6 in 10 of the 5 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

Tracker

Format and status values: shared/references/tracker.md

DATA_DIR/tracker.md is the single source of truth. This skill reads it, verifies it against the job folders, and tells the user what needs doing.

Step 1: Load and verify

Read tracker.md and list DATA_DIR/jobs/. Run the integrity checks from the reference:

  • Every job folder has a row, every row points at a folder that exists
  • No duplicate rows for the same employer plus role
  • No row with Applied set but status still Bewertet
  • Scores match the evaluation.md in each folder
  • No row past its Next date without being surfaced

Report discrepancies, do not silently repair them. Show the row and what is wrong, then propose the fix. Data the user might care about is not yours to quietly rewrite.

Step 2: Present the pipeline

Funnel first, then what needs action, then the table:

Pipeline — [date]

  Gefunden      12
  Bewertet      31   (avg score 3.4)
  Gesendet      18
  Gespräch       4
  Angebot        1
  Absage         9
  Verworfen     14

Response rate: 22% (4 Gespräche from 18 Bewerbungen)

Needs action today
  · Beispiel GmbH — follow up due (sent 3 weeks ago, Mittelstand)
  · Muster AG — Gespräch Thursday 14:00, no prep done
  · Vorbild SE — evaluated 4.5, never applied

Then the filtered table. Default to open items — hide Absage and Verworfen unless asked.

Step 3: Flag the stale

Using the German follow-up timing table (startups ~10 days, Mittelstand ~3 weeks, Konzern ~4 weeks, öffentlicher Dienst ~6 weeks), list anything overdue and offer /wingman:followup.

Remember: an Eingangsbestätigung is not a response, and a Zwischenbescheid means the process is still running — neither is a rejection.

Step 4: Offer next actions

Point at the specific command for each item needing attention. Never do the work here.


Response Format

  1. Funnel — counts by status, response rate
  2. Needs Action — overdue follow-ups, upcoming interviews, evaluated-but-not-applied
  3. Integrity — discrepancies found, or clean
  4. Pipeline — the table

Permissions

{ "permissions": { "allow": [
  "Read(~/.wingman/**)", "Write(~/.wingman/**)", "Edit(~/.wingman/**)"
] } }

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.