aksheyw
UserPre-push quality gate for Claude Code that runs above git, so git push --no-verify can't skip it. Tests, code review, security, and a secret scan must pass before any push to your protected branch. Say 'ship it' to run it.
Categories
Indexed Skills (26)
ship-init
Scaffold a .shipgate.json for this project by auto-detecting its test/lint/typecheck/build commands and deploy target.
ship
Run the ship-gate quality gates and push to the protected branch (usually main). Invoke when the user expresses intent to ship, release, or push to the main/protected branch (e.g. "ship it", "go ahead and ship", "sync with GitHub"). On an imperative command, auto-push when all gates pass; on a question or a "not yet", run the gates and report without pushing.
llm-wiki
Persistent project wiki that compounds knowledge across sessions. Ingest learnings, query past decisions, lint for contradictions. Based on Karpathy's LLMwiki pattern.
apply
Paste a JD to get full analysis + tailored resume + cover letter. Argument: [paste JD text here]
evaluate
Evaluate a JD before applying — 7-block assessment with go/no-go score. Argument: [paste JD text or URL here]
followup
Follow-up cadence tracker. Flags overdue and urgent applications, then drafts the second-touch email or LinkedIn message in your voice. Argument: none, or a company name to follow up on one.
humanizer
Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, negative parallelisms, and excessive conjunctive phrases.
interview
Full interview prep for a specific company. Argument: [company name] [optional: role title]
outreach
Generate the right outreach message for any contact or company. Argument: [person name / company / context — e.g., 'alumni at Google' or 'recruiter at Stripe']
patterns
Analyze the application tracker for what converts and what wastes time. Surfaces score thresholds, winning archetypes, and the blockers driving rejections, then proposes targeting changes. Argument: none.
resume-customizer
Director-level / senior PM resume + cover letter toolkit. Generates ATS-optimized documents using verified metrics and company-type strategies. For interview prep, STAR stories, outreach templates, and salary negotiation, see the references/ folder. Personalize `references/YOUR_PROFILE.md` first.
salary-negotiate
Full salary negotiation prep — current comp, target, scripts, and counter-offer handling. Argument: [company name] [optional: offer details]
track-application
Log a new application or check pipeline status. Argument: [company] [role] [status: applied/screening/interview/offer/rejected]
deep-review
Iterative deep review using 14-lens methodology. Use when reviewing ANY plan, code review, architecture doc, test plan, or security assessment — before claiming it is complete. Runs multiple passes from different analysis angles until no new findings emerge. Each lens catches issues invisible to previous lenses. Found 14 production bugs (2 ship-stoppers) in its first use. ALWAYS use this when the user asks to review something thoroughly, verify completeness, audit a document, or check if anything was missed. Also triggers on "did you miss anything", "is this thorough", "are you sure", "check again", or similar phrases.
ship-review
Ship Gate code-review gate, runs /code-review then applies the distilled checklist and emits Approve/Warning/Block.
generic-letter
Generate the network forwarding cover letter for friends, alumni, or general network outreach.
update-memory
Log what worked (or didn't) to improve future applications. Argument: [company] [outcome] [what worked / what to change]
docker-config-qa-regression
Systematic verification checklist after modifying Docker container config (volumes, env vars, compose edits). Use after any docker-compose change to catch cascading failures (auth loss, ephemeral volumes, missing credentials) before they reach production. Originally derived from a real session debugging post-config-change auth failures.
docker-ephemeral-volume-fix
Diagnose and fix the "settings randomly reset" pattern in Docker containers — when an app migration writes active config to a new path that isn't volume-mounted, dashboard/UI changes silently disappear on `docker compose down/up`. Includes diagnostic checklist and the copy-out-then-mount fix.
openai-compatible-embeddings-batching
Custom embeddings adapters against OpenAI-compatible endpoints (OpenRouter, vLLM, Together, Anyscale, Groq, self-hosted gateways) must batch transparently and fail loudly — some providers return HTTP 200 with a non-standard error body missing the `data` key, which surfaces as a bare KeyError deep inside LangChain. Includes the two-layer adapter fix and the regression test that actually exercises the batching path.
prompting-opus-4-8
How to prompt and scaffold Claude Opus 4.8 — effort/thinking defaults, verbosity, literal instruction-following, tool-use and subagent triggering, code-review recall, and the cream/serif frontend house-style that's wrong for dashboards. Use when starting Opus 4.8 work, migrating prompts from Opus 4.7, tuning verbosity or effort, wondering why the model isn't "thinking", seeing a code-review recall drop, or getting a cream+serif design you didn't ask for.
prompting-qwen3-coder
How to prompt and run Qwen3-Coder (qwen3-coder:30b via Ollama) — a NON-thinking model with no deliberation channel, optimised as an agentic patch-writer (though it CAN critique — measured). Covers the official sampling params, tool calling as the vendor's structured path, the 256K/YaRN context story, and the "silent or flooding" failure signature — including the measured finding that a reviewer prompt's own suppression clauses, NOT grammar-forced JSON, silence it. Use when prompting Qwen3-Coder, wiring it into a local council/review seat, debugging an empty or fabricating seat, tuning sampling, or deciding whether it is the right model for a job.
prompting-sonnet-5
How to prompt and scaffold Claude Sonnet 5 — the migration-from-4.6 API breakers (sampling params now error, new tokenizer inflates tokens ~30%, extended-thinking budgets removed), adaptive-thinking-on-by-default, effort mapping, tool-use triggering, and design/code-review tuning. Use when starting Sonnet 5 work, picking Sonnet as a subagent model, migrating a 4.6 prompt or n8n/API call, hitting a 400 error on temperature, or seeing max_tokens truncation.
audit
Comprehensive codebase audit skill covering 10 systemic bug categories plus CI automation. Use this skill proactively during code review, E2E testing, pre-deployment checks, or whenever you spot any of these patterns — raw database values in UI (snake_case, camelCase enums, numeric codes), API contract mismatches between client and server, Firestore/database rules gaps, CORS missing production domains, analytics/consent gating features, dynamic Tailwind/CSS classes purged in prod, window.open without noopener, orphaned data on deletion, hardcoded secrets, Android/iOS security misconfigs, performance regressions (bundle size, re-renders, lazy loading), or accessibility gaps (missing aria labels, keyboard nav, screen reader). Also triggers on "audit this", "did you check everything", "is this production ready", "what did we miss", "are you sure", or any thoroughness question. One bug instance always means many more exist — this skill enforces the ripple search.
context-bridge
Use when starting, continuing, or closing a Claude Code session on a multi-session project. Scaffolds a small per-project wiki (.claude/wiki/) and runs an 11-step save+sync that produces a paste-and-resume handoff prompt. Solves cross-session amnesia. Does NOT solve mid-session context bloat — use /strategic-compact for that.
ship-security
Ship Gate security gate, applies the distilled security checklist to the diff; optionally augments with /security-review.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.