| name | finish-app-script |
| description | Drain Liam's tailored-resume application queue from a phone-friendly command. Build `/tmp/fa_script_run_state.json`, then run the rotating batch orchestrator that launches fresh Codex CLI parent processes for small Chrome/Computer Use batches, closes submitted tabs, leaves medium/low-confidence handoff tabs open, updates tracker/cache, and commits confirmed submissions. Use when the user says `$finish-app-script`, "run finish app script", "drain my application queue", or wants applications completed without context overflow. |
Finish-App-Script
A reliability-focused parallel of finish-applications. Same goal — submit ready tracker rows — but uses an outer orchestrator to rotate fresh Codex CLI parent processes through small batches.
Default Invocation
When the user invokes $finish-app-script or asks to run this skill from chat,
do this without asking for extra confirmation:
python3 skills/finish-app-script/scripts/run_monitored_batches.py
Use run_monitored_batches.py by default for live browser work. It refreshes
the tracker cache, builds /tmp/fa_script_run_state.json, runs run_batches.py,
and restarts the batch runner if it stops while queued rows remain. Monitor its
terminal output while it runs. If the assistant turn is interrupted or the shell
session disappears, inspect /tmp/fa_script_run_state.json; when queued rows
remain and no run_batches.py/codex exec process is active, resume with:
python3 skills/finish-app-script/scripts/run_monitored_batches.py --resume
Do not pass --no-commit or --no-push unless the user explicitly asks.
run_batches.py owns commits and pushes, including final commits for confirmed
submissions. Use the legacy run_queue.py only if the user explicitly asks for
per-row mode.
Default queue behavior includes all not-applied rows that pass the script's
basic filters, including rows previously labeled true manual or Workday. Those
rows are queued as low-confidence live-browser attempts so the agent can fill
safe fields and answerable FRQs, then leave the tab open with a precise blocker
or review note. Final submission is still allowed only for high-confidence rows.
If Chrome/Computer Use times out or is unavailable, stop the run as a systemic
browser blocker instead of marking the rest of the queue manual. If Firefox is
responsive through Computer Use, future batches may use Firefox as the fallback
browser so the queue can continue without losing the high-confidence-only
submission gate.
Email 2FA, verification codes, and magic links sent to liamvanpj@gmail.com are
not blockers; the batch agent should use Gmail access to retrieve the code or
open the link and continue. Only SMS/authenticator-app 2FA, missing/expired
emails, or account-creation/legal gates should become manual blockers.
For native macOS file uploads, use Cmd+Shift+G with the exact absolute PDF path
from the row state, then Return/Open, and verify the rendered attached filename.
Do not trust folder-click navigation when the picker remembers a previous
application's directory or stale file.
For Greenhouse upload widgets, click the nested Browse... button inside the
Resume/CV field rather than the outer Attach control; the
outer control can leave Open disabled even when a PDF is selected.
If Firefox selects the exact existing PDF but leaves the native picker Open
button disabled, treat that as a Firefox upload bug, not a missing-file blocker:
retry the same public ATS form in Safari using the same nested Browse... and
Cmd+Shift+G exact-path flow before marking the row manual.
For known upload-redo rows whose notes or blocker mention Document upload failed, upload-error redo, or Firefox picker, start the document upload in
Safari instead of spending the retry on Firefox. Safari has been verified on
Uare.ai Greenhouse for resume PDFs.
Tab Confidence Grouping
During live browser runs, keep open application tabs organized by perceived
handoff confidence:
- High Confidence / Ready Submit: fully completed tabs where every required
answer is covered by standing answers, critical rendered answers are verified,
and only final submission/confirmation remains.
- Needs Review: tabs with one or more medium-confidence answers, free
response drafts, eligibility/location/salary uncertainty, or required choices
that Liam should inspect.
- Hard Blocker: Workday, login/account creation, SMS/authenticator 2FA,
CAPTCHA, legal signature/attestation, AI-deterrent verification, or other true
blockers.
- Submitted / Archived: confirmation tabs, already-submitted portals, closed
postings, mismatched redirects, and not-found pages.
If Chrome tab groups are available through UI automation, place tabs into those
groups. If tab groups are not scriptable from the current environment, use the
closest non-destructive fallback: reorder tabs into the same sequence and keep a
short written map in the run notes or final response.
How To Run
python3 skills/application-visualizer-refresh/scripts/refresh_visualizer_data.py
python3 skills/finish-app-script/scripts/build_queue.py
python3 skills/finish-app-script/scripts/run_monitored_batches.py
python3 skills/finish-app-script/scripts/run_monitored_batches.py --resume
python3 skills/finish-app-script/scripts/run_batches.py
python3 skills/finish-app-script/scripts/run_queue.py
run_batches.py flags:
--batch-size N — rows per fresh Codex process (default 2)--max-batches N — stop after N fresh Codex processes (testing)--model MODEL — override default gpt-5.5--timeout SECONDS — per-batch timeout (default 1800s = 30 min)--child-sandbox MODE — sandbox for each fresh Codex process (default danger-full-access)--no-commit — skip auto-commit every 5 confirmed submissions--no-push — commit but don't push--dry-run — print what would be spawned without invoking codex exec
run_monitored_batches.py flags:
--resume — skip refresh/build and continue /tmp/fa_script_run_state.json--max-restarts N — retry stopped runs with queued rows remaining (default 3)--restart-sleep SECONDS — wait before retrying (default 10)- all normal
run_batches.py flags above are forwarded
run_queue.py flags:
--max-rows N — stop after N rows (testing)--model MODEL — override default gpt-5.5--timeout SECONDS — per-row timeout (default 360s = 6 min)--no-commit — skip auto-commit every 5 confirmed submissions--no-push — commit but don't push--dry-run — print what would be spawned without invoking codex exec
Architecture
Rotating Parent Mode (run_batches.py)
run_batches.py (outer orchestrator)
├── reads /tmp/fa_script_run_state.json
└── while queued rows remain:
├── previews the next N queued rows (default 2)
├── spawns a fresh parent Codex CLI process:
│ codex exec --ephemeral --cd $REPO --sandbox danger-full-access
│ -m gpt-5.5 -o /tmp/fa_script_batch_outputs/batch_NNN.txt "<prompt>"
├── that fresh parent owns Chrome/Computer Use directly
├── parent processes up to N rows, writes outcomes to state, exits
├── outer orchestrator re-reads state and summarizes progress
└── commits + pushes every 5 confirmed submissions
This is the preferred live-browser mode. It keeps the context-reset benefit,
but avoids making every individual application a tiny GUI worker. Each fresh
Codex process behaves like a short $finish-applications run: it owns the
browser, processes one row at a time, and stops after a small batch.
Legacy Per-Row Mode (run_queue.py)
run_queue.py (orchestrator)
├── reads /tmp/fa_script_run_state.json
└── for each queued row:
├── builds tight per-row prompt (~250 tokens)
├── spawns: codex exec --cd $REPO --sandbox workspace-write
│ -m gpt-5.5 -o /tmp/fa_script_outputs/<key>.txt "<prompt>"
├── waits up to 6 min
├── re-reads state file → checks outcome (submitted|manual|archived)
├── increments counters, runs circuit breaker
└── commits + pushes every 5 confirmed
Each spawned agent is a fresh process with no context history. It reads OPERATING_CARD.md, applies to one job, writes outcome to the state file, exits.
Why This Architecture
| Pain point | Why this fixes it |
|---|
| Context overflow | Rotating parent mode sees only a small batch before the CLI process exits |
| Computer Use flakiness | The browser is owned by a fresh parent process for a short batch, instead of a separate GUI worker per row |
| FRQ blocking the queue | Fill-and-leave-open enforced per row; orchestrator continues |
| Single-agent drift | The submission rule lives in the prompt itself, can't degrade |
Bug Fixes vs finish-applications
build_queue.py corrects three behaviors in the original build_application_queue.py:
"not automation-accessible" is retryable, not a true blocker. Past notes from environment failures (denied Chrome automation permission) used to permanently lock those rows; they re-enter the queue now.- Manual-Apply-Needed rows with no specific note are retryable. Original
true_manual_reason() returned a generic non-empty fallback that forced action=manual. Returns "" here so they get action=retry/apply.
- FRQ terms reduce confidence, not block the row.
"free response", "free-response", "custom" moved out of TRUE_MANUAL_BLOCKERS into a new CONFIDENCE_REDUCERS list (-20 score impact, pushes to medium band so the agent attempts with fill-and-leave-open).
Per-Row Prompt Template
The orchestrator builds this for each spawned agent. Keep it tight; do not bloat with full SKILL contents — the agent reads OPERATING_CARD.md for rules.
Read skills/finish-app-script/OPERATING_CARD.md before starting. Follow it strictly.
You are completing ONE job application.
Company: {company}
Role: {role}
URL: {jobLink}
Resume PDF: {resumePdf}
Posting key: {postingKey}
Source: {source}
State item key: {key}
Notes carried from tracker: {notes}
Use Codex Computer Use for the browser. Dropdowns/typeaheads: open menu →
click option → verify chip rendered. Never just type and move on.
Email 2FA / verification codes / magic links to liamvanpj@gmail.com are NOT
blockers. Use gmail@openai-curated MCP to read the code or click the link in
Chrome (already signed in), then continue. Only escalate if the email never
arrives, expires, or the verification switches to SMS / authenticator-app 2FA.
Confidence decision after final review:
- HIGH (every required field covered, no blocker): click submit, retrieve any
emailed verification code via Gmail MCP, capture confirmation,
run update_application_status.py, set state="submitted".
- MEDIUM (FRQ/one uncertain field): fill all safe fields, generate best-effort
answer from profile, leave tab open, set state="manual" with
blocker like "FRQ review: <question>".
- HARD blocker (account creation, fresh login when unauthenticated, SMS/
authenticator-app 2FA, interactive CAPTCHA, Workday, legal signature):
set state="manual" with exact blocker.
- Posting closed/404/redirected: set state="archived".
Update /tmp/fa_script_run_state.json item with key "{key}":
state, result, blocker (if manual), confirmationEvidence (if submitted), updatedAt.
Exit when done.
Run State File
/tmp/fa_script_run_state.json (separate namespace from the legacy skill's /tmp/fa_run_state.json). Schema mirrors the legacy file but items[i].state transitions are the only thing the orchestrator reads back.
Liam's Standing Answers
Detailed in OPERATING_CARD.md rule 5. The condensed list is in the per-row prompt only as a reference — the agent re-reads the operating card on every run.
Factual Resume Context
Every spawned agent must use Liam's real work history as the factual basis for
FRQs, values answers, achievement examples, and project examples. The row's
tailored resume.tex next to resumePdf is the
immediate source of truth. If generic-resume/README.md or
generic-resume/resume.tex exists, the agent must read those too and treat them
as the broader evidence bank. Never invent employers, internships, tools,
projects, metrics, dates, credentials, or responsibilities. If a requested
answer cannot be grounded in the resume/profile/tracker evidence, the agent
should use a supported adjacent example or leave the tab open for review.
Education dates are strict: University of Georgia BS Computer Science, started
Aug 2021, graduated Dec 2024. Never enter a graduation year before 2024, and
never answer that Liam graduated before 2020. If a form asks whether Liam
graduated before 2020, answer No.
Cover Letters
Do not generate, render, write, paste, or upload cover letters. If a
cover-letter field is optional, leave it blank. If a cover-letter field is
required and cannot be skipped, leave the tab open and mark the row manual with
Cover letter required; skipped by no-cover-letter policy.
Workday
Workday rows are flagged at queue-build time and never spawned — Liam submits manually. To mark them in the tracker:
python3 skills/finish-app-script/scripts/build_queue.py --mark-workday-manual
Circuit Breaker
run_queue.py stops the run if 3 consecutive rows return state=manual. This catches Chrome auth loss, network problems, or systemic blocker patterns before burning more spawns. Resume by clearing the consecutive-manual streak (run with --max-rows to test) or fixing the underlying issue.
Final Response
Orchestrator prints a summary: confirmed submissions, manual rows with blockers, archived rows, commits made. Tracker and visualizer cache reflect all confirmed submissions.
