AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Application Tracker

skill-bydeng01-phd-application-skill-application-tracker · by bydeng01

>-

No reviews yet
0 installs
6 views
0.0% view→install

Install

$ agentstack add skill-bydeng01-phd-application-skill-application-tracker

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-bydeng01-phd-application-skill-application-tracker)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
1mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Application Tracker? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Application tracker

A PhD applicant is running many parallel processes — different deadlines, document checklists, and pending replies across a dozen labs — and the failure mode is silent: a missed deadline or a follow-up that never happened costs an entire opportunity. This skill is the applicant's air-traffic control. Its value is surfacing the few things that need action now out of a mass of state, accurately, so nothing winnable slips.

What you produce

An updated set of status.md files plus a concise action digest for the user: what's overdue, what's due soon, what's missing, and what's waiting on someone else. Accuracy and prioritization matter more than completeness — a digest that buries the one urgent deadline under thirty lines of "no action needed" has failed.

Step 1 — Gather state

Read every knowledge-base/applications/*/status.md (the stage, deadline, priority, checklist, and timeline) and the knowledge-base/interactions/*.md logs (what was sent, when, and any Follow-up due dates). Determine today's date with the shell rather than assuming it — overdue/upcoming calculations depend on it.

Step 2 — Compute what needs attention

Sort everything into a small number of action buckets, ordered by urgency:

  • Overdue — a passed application deadline still marked unsubmitted, or a Follow-up due

date in the past with status still "awaiting reply". These are the headline items.

  • Due soon — deadlines or follow-ups within the next ~14 days. Show the date and the

days remaining so urgency is legible.

  • Missing documents — checklist items still pending for applications whose deadline is

approaching (transcripts, letters, writing samples, test scores). Surface these early because documents like recommendation letters have their own lead time.

  • Waiting on others — sent emails or submitted applications where the ball is in the

professor's or committee's court; note when a polite follow-up becomes appropriate.

  • No action needed — everything else, summarized in a single line, not enumerated.

For follow-up timing, respect the dates logged by outreach-email; a cold email typically warrants a follow-up ~2 weeks out, and only one or two follow-ups before moving on.

Step 3 — Update the records

Where the state has clearly changed, update the status.md files: advance stage, tick completed checklist items, correct deadlines you've confirmed. Keep edits faithful to what you actually know — don't mark something submitted on assumption. If you change a record, note it so the user can see what moved.

Step 4 — Deliver the digest

Lead with the action buckets, most urgent first. Be specific and scannable:

## PhD application status — 

### ⚠ Overdue / act now
- : follow-up to Prof X is 5 days overdue (emailed , no reply) → send follow-up
- ...

### Due within 14 days
- : application deadline  (in 9 days); still missing: writing sample

### Waiting on others
- : submitted ; committee decisions typically 

### On track
- 6 other applications, nothing due in the next two weeks.

End with a short recommended next-action list tied to the urgent items (e.g. "draft the overdue follow-up to Prof X" → hand to outreach-email; "request the missing transcript").

Step 5 — Offer recurring tracking

Staying on top of deadlines is inherently recurring, and the whole point is to catch things before they're missed. Offer to set up a scheduled task (e.g. a Monday-morning digest) that re-runs this check and surfaces only what needs action that week. Many applicants don't realize this can run automatically.

Guardrails

Never invent or guess a deadline or a status — an applicant who relies on a fabricated date is worse off than one who knows the date is unconfirmed. When a date or requirement is unverified, say so and flag it for the user to confirm at the source. Don't auto-send follow-ups; surface that one is due and hand the drafting to outreach-email for the user to review and send.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.