AI Agent Workflow Collection

Application Tracker — manages JOB- prefixed items in ~/.personal/tracker.json. Aggregates application status, surfaces stale items, generates weekly hunt reports, integrates with the dashboard. Triggered by 'where am I on the hunt', '/hunt-status', 'weekly summary', or auto-runs Sunday evenings.

Periodic asks-vs-state audit. Extracts Josh's explicit asks from session DB, reconciles against current state across all projects/skills/repos. Surfaces drift between 'what Josh asked for' and 'what actually happened'. Runs every 14 days via launchd or on demand via /audit.

Company Research — produces a cached per-company brief covering recent news, leadership, funding, eng-blog signal, Glassdoor read, podcast appearances, tech stack, public talks. Tier 2 sub-agent does the fetch, Tier 3 main synthesizes. Reused across multiple applications at same company.

Content idea bank manager — captures, mines, and ranks post ideas for LinkedIn and Substack. Three modes -- /idea for instant capture (5s), weekly mine for source-sweep (4 parallel sub-agents over project-logs / OpenCode session DB / Obsidian vault / readings), and /content-rank for "what should I post next." Outputs ~/.personal/content-bank/ideas.json. Skips ideas that overlap recent posts.

Polish-only editor for an existing draft. Removes AI tells, enforces platform rules, preserves voice. Output is a diff (what changed and why), not a wholesale rewrite. Tier 2 main session. Triggered by 'edit this draft', 'polish this post', '/edit-draft [POST-NNNNN]'.

Repurposes an existing post across platforms with full voice recalibration -- not copy-paste. Substack-long → LinkedIn excerpt+link, or LinkedIn → Substack short newsletter. Reads source post, extracts the sharpest standalone claim, redrafts in target platform's voice. Tier 3 main session. Triggered by 'repurpose POST-NNNNN', '/repurpose [POST-NNNNN] [target-platform]'.

Cover Letter — drafts a voice-matched, JD-tailored cover letter capped at 350 words. Tier 4 main session. Anti-AI-tells gates. Triggered by 'cover letter for [company]', '/cover-letter', or after resume tailoring completes.

Audit a repo's docs for drift between README, AGENTS.md, code, declared paths, and reality. Surfaces stale references, broken links, placeholder text, missing required files. Reports Red/Yellow/Green per layer with file:line evidence; never auto-fixes.

Interview Prep — generates a structured pre-round brief: predicted questions by round type, STAR story mapping, three sharp questions to ask, follow-up email skeleton. Tier 4 main session for the brief itself, Tier 2 sub-agents for company/role research.

Job Description Analyzer — the engine of the filter-don't-spray pattern. Reads a JD URL or pasted text, scores it A-F across 6 axes (archetype/fit/positioning/comp/tailoring/interview), produces a structured fit report and tailoring plan. Skip anything <4.0. Triggered by 'analyze this job', 'is this worth applying to', 'score this JD', or any URL pointing at a job listing.

Drafts a LinkedIn post from an idea ID + voice profile + platform rules + recent-posts memory. Produces 3 variants with different hook strategies. Voice smell test (em-dash ban + AI-tells grep) gates output. Tier 3 main session. Triggered by 'draft a LinkedIn post', 'LinkedIn this idea', '/linkedin [IDEA-NNNNN]'.

Mock Interviewer — adaptive role-play sub-agent that asks behavioral / technical / system design questions, scores answers against rubrics, feeds findings back into the STAR bank. Tier 3 main (Sonnet) for the dialogue. Triggered by 'mock interview', 'practice interview for [role]', '/mock'.

Networking Outreach — drafts warm intros, cold DMs, recruiter responses, post-interview follow-ups. Voice-strict (Tier 4 Opus). Hard caps on length per channel. Anti-AI-tells gates. Triggered by 'draft a DM to [person]', 'reach out to [person]', 'recruiter responded — what do I say', '/outreach'.

Pre-push adversarial review for personal repos before git push. Runs adversarial-input sweep, parallel-implementation symmetry audit, project AGENTS.md compliance, dead-code sweep, secrets scan. Use before any push; mandatory for repos that will become public or shared. Tier 2 sub-agent does the bulk; Tier 3 main synthesizes.

Provider Router — answers "which model should I use for X", "what's the cheapest way to do Y", "is there a free option for Z" by walking the kit's routing matrix. Triggered by model-selection questions, cost-comparison questions, or when the user asks for guidance on multi-provider routing. Returns a recommendation with cost estimate and the agent name to invoke.

Bootstrap a new personal repo with industry-best-practice docs and defensive .gitignore. Routes by repo type (scratch / personal-published / portfolio-public). Ensures secrets/PII protection from the first commit. Triggered by 'create new repo', 'bootstrap [name]', '/repo-bootstrap'.

Resume Tailor — produces an ATS-optimized, JD-aligned resume from master CV + JD analysis. Two-stage pattern (Tier 2 sub-agent extracts JD keywords, Tier 4 Opus drafts). Outputs markdown + Reactive Resume PDF. Triggered by 'tailor resume for [company]', '/tailor-resume', or after JD analysis returns RECOMMEND APPLY.

Salary Negotiation — triggered at offer time. Pre-built scripts for common scenarios (geo-discount pushback, RSU acceleration, competing-offer leverage, signing-bonus ask, base-vs-equity tradeoffs). Tier 4 main session. Voice-strict.

Recursive self-healing framework for skill improvement. Scores executions, detects failures, proposes auto-repairs, tracks maturity progression. Triggered by 'why did [skill] fail', 'improve [skill]', '/self-heal [skill]', or auto when a skill's invocation produces an explicit Josh-correction.

Session Handoff — appends a session entry to this project's project-log.md, then prints a copy-paste prompt for the next session that points at that log. Use when wrapping up a session that crossed turn 40+ or to end clean before doom-loop territory. Triggered by 'wrap up', 'handoff', '/handoff'.

Standards and quality gates for the skill library. Audits existing skills (descriptions, dependencies, dead skills, eval coverage), proposes improvements, governs how new skills are created. Runs on demand or quarterly. Triggered by 'audit skills', 'skill health check', '/skill-maint'.

STAR Story Extractor — Josh's unfair advantage. Mines OpenCode session DB, project-logs, account memory, meeting transcripts, prior work artefacts to extract Situation/Task/Action/Result/Reflection stories for behavioral interview prep. Multi-source delegated extraction. Builds the star-bank.json that feeds resume-tailor and cover-letter skills.

Drafts a Substack piece from an idea ID + voice + platform rules + STAR bank for grounding. Two modes -- long (1000-2500w, Tier 4 Opus + adversary pass) and short (300-800w newsletter, Tier 3 Sonnet). Title + subtitle craft block mandatory. Voice smell test gates output. Triggered by 'draft a Substack piece', '/substack [IDEA-NNNNN]', '/substack-short [IDEA-NNNNN]'.

Weekly tooling discovery. Scans OpenCode docs/changelog, awesome-opencode, npm opencode- packages, and major MCP servers for new releases worth trying. Outputs a Monday-morning summary. Triggered by '/tooling-watch' or auto Monday 8am via launchd.