Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/suxrobgm/jobpilot/pilotnpx skills add suxrobGM/jobpilot --skill pilotgit clone --depth 1 https://github.com/suxrobGM/jobpilotWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00033 | $0.07618 |
| Opus 5 | $0.00016 | $0.03809 |
| Sonnet 5 | $0.00007 | $0.01524 |
| Haiku 4.5 | $0.00003 | $0.00762 |
Grade A, and why
pilot scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -fsS -H "authorization: Bearer $JOBPILOT_API_TOKEN" "$JOBPILOT_API/api/pilot" How it starts
The opening of the file, as written. The whole thing — 372 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pilot - One Autonomous Cycle
JobPilot's autonomous mode: the host re-injects this skill perpetually, so each invocation is one stateless cycle - sense, decide, act, record, exit. All state lives in the API; nothing survives between invocations except what you write there. Do exactly one agenda item (at most one worker delegation, one browser activity), journal it, print the sentinel, stop.
0. Setup
Follow ../_shared/setup.md - health check GET /api/health first; abort with its standard message if down. Then generate a cycle id (uuidgen if present, else a portable fallback):
CYCLE_ID=$(uuidgen 2>/dev/null || cat /proc/sys/kernel/random/uuid 2>/dev/null || od -An -tx1 -N16 /dev/urandom | tr -d ' \n' | sed -E 's/^(.{8})(.{4})(.{4})(.{4})(.{12})$/\1-\2-\3-\4-\5/')
Load the pilot state - step 2 breaks priority ties with its goals text. No run-state check here: the host gates the loop.
curl -fsS -H "authorization: Bearer $JOBPILOT_API_TOKEN" "$JOBPILOT_API/api/pilot"
1. Sense
AGENDA=$(curl -fsS -H "authorization: Bearer $JOBPILOT_API_TOKEN" -X POST "$JOBPILOT_API/api/pilot/agenda/refresh")
AGENDA_VERSION=$(echo "$AGENDA" | jq -r '.version')
A 409 means the pilot was stopped mid-cycle - a rare race the host normally gates. Journal and exit empty.
curl -fsS -H "authorization: Bearer $JOBPILOT_API_TOKEN" -X POST "$JOBPILOT_API/api/pilot/journal" \
-H 'content-type: application/json' \
-d "$(jq -n --arg cid "$CYCLE_ID" \
'{cycleId:$cid, entries:[{kind:"cycle", summary:"Pilot is stopped.", detail:{status:"empty", sleepSeconds:3600}}]}')"
Print [[JOBPILOT_CYCLE cycle=$CYCLE_ID status=empty sleep=3600]] as the final line, stop.
If .items is empty:
curl -fsS -H "authorization: Bearer $JOBPILOT_API_TOKEN" -X POST "$JOBPILOT_API/api/pilot/journal" \
-H 'content-type: application/json' \
-d "$(jq -n --arg cid "$CYCLE_ID" --arg s "All caught up - nothing needs doing; checking back at <nextWakeAt>." \
--argjson detail "$(echo "$AGENDA" | jq '{status:"empty", sleepSeconds:.sleepSeconds}')" \
'{cycleId:$cid, entries:[{kind:"cycle", summary:$s, detail:$detail}]}')"
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 372 lines · 33 tokens per session scan A 70e8fe6c24c2
pilot is a skill published in the GitHub repository suxrobGM/jobpilot (60 stars, last pushed 11d ago), licensed MIT. It adds 33 tokens to every session and 7,618 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
scrape
Finds new job postings matching your profile via installed portal-search CLIs (job-board APIs and any skills added with /add-portal). Deduplicates across runs. Triggers on: job scrape, find jobs, search jobs, new jobs, job search, scrape jobs, /scrape.
scrape
Finds new job postings matching your profile via installed portal-search CLIs (LinkedIn, local job boards, and any skills added with /add-portal). Deduplicates across runs. Triggers on: job scrape, find jobs, search jobs, new jobs, job search, scrape jobs, /scrape.
job-application-agent
Finds, evaluates, fills, submits, and tracks a candidate's own job applications using a verified resume, evidence-based targeting, secure local profile storage, and browser automation. Use for onboarding or migrating a job-search profile, searching active roles, assessing a posting, applying to an authorized URL or…
job-apply
Fill out job applications automatically using your resume. Use when the user wants to apply for jobs on LinkedIn Easy Apply, Greenhouse, Ashby, Lever, Rippling, or Workday.
answer-memory
Manage Job Apply's local profile, reusable answers, application history, and resumable sessions. Use whenever a Job Apply workflow needs to initialize, migrate, read, or update persistent applicant data.
job-search
Search LinkedIn, Hacker News, and Twitter/X for jobs with connections, hiring manager insights, and preference-based scoring. Use when the user wants to find jobs, search for positions, or explore job opportunities.