skyfox675

@skyfox675 source repo

30 skills
0 followers
4 hours ago last updated

▌
PDF · skyfox675 bundle
Use this skill whenever the user wants to do anything with PDF files. This includes reading or extracting text/tables from PDFs, combining or merging multiple PDFs into one, splitting PDFs apart, rotating pages, adding watermarks, creating new PDFs, filling PDF forms, encrypting/decrypting PDFs, extracting images, and OCR on scanned PDFs to make them searchable. If the user mentions a .pdf file or asks to produce one, use this skill.

0 installs
▌
DOCX · skyfox675 bundle
Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.

0 installs
▌
PPTX · skyfox675 bundle
Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions "deck," "slides," "presentation," or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.

0 installs
▌
XLSX · skyfox675 bundle
Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.

0 installs
▌
Figma · skyfox675 bundle
Connect to Figma from your AI tool — read design context (layout, typography, colors), fetch variables/tokens, capture visual references, generate code from frames, and optionally map Figma components to code with Code Connect. Use whenever a task involves a Figma file, design tokens, a design-to-code translation, or the Figma MCP. This repo's figma-specs skill and /figma-init-spec / /figma-refresh-spec commands build on it (Figma MCP read tools required; Code Connect optional).

0 installs
▌
Caveman · skyfox675 bundle
Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.

0 installs
▌
Humanizer · skyfox675 bundle
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, passive voice, negative parallelisms, and filler phrases.

0 installs
▌
Pr Checks · skyfox675 bundle
Single-lane interval watcher that keeps open agent PRs green and mergeable — the CI + conflict + merge-queue medic lane of a multi-agent fleet. Watches for failing required checks, DIRTY/conflicted branches, AND merge_group (queue) failures that head checks are blind to; dispatches fixes, dequeues + drafts poison-pill PRs that drop a whole ALLGREEN group, and re-arms innocent PRs ejected by a group drop. Loops on an interval ("/loop <interval> start"). Use when running a CI/checks lane, when a PR is green on its own head yet never merges, when one bad PR stalls the entire merge queue, or when telling a real poison pill apart from a flake or an innocent ejection.

0 installs
▌
Pr Cleanup · skyfox675 bundle
Single-lane interval watcher that tidies up after agent PRs close — the post-merge janitor lane of a multi-agent fleet. Acts ONLY on closed PRs — deletes merged local branches, removes spent worktrees, closes the linked GitHub/Jira issue on merge, and releases the claim lock. Critical when PRs land on an integration branch (not the repo default), because GitHub then does NOT auto-close "Closes

0 installs
▌
Figma Specs · skyfox675 bundle
Tame design drift with the Figma MCP — snapshot a Figma file into living design specs anchored to STABLE identity (node IDs and published component/style/variable keys) rather than page/frame names or position, so when designers reorganize the file the links still resolve. Each component is mapped to its code by a repo-owned design→code map (discovered by name-matching the codebase and confirmed once; Code Connect is an optional accelerator, never required). Refresh re-reads Figma, diffs against the snapshot by stable id, and emits a drift report that separates pure reorganization (no code change) from real token/size/state changes (a precise re-touch list pointing at the mapped code) — so tiny differences get caught and fixed once instead of accumulating. Use when designs or the Figma file organization keep changing and the code keeps needing re-touches, to establish or refresh the design spec set, or when asked to track Figma changes, detect design drift, or sync design to code. Domains are discovered from

0 installs
▌
MCP Builder · skyfox675 bundle
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).

0 installs
▌
Pr Comments · skyfox675 bundle
Single-lane interval watcher that drives PR review threads to resolved, then arms auto-merge — the review-thread-closer lane of a multi-agent fleet. Watches every open agent PR for AI-review-bot and human review comments, dispatches a fix sub-agent or posts a reasoned counter, resolves the thread only after the fix is pushed, and owns the INITIAL arm-after-bot-review (deliberately not done at dispatch, to avoid racing the bot). Loops on an interval ("/loop <interval> start"). Use when running a comments/review lane, when reviewed-but-unmerged PRs pile up because unresolved threads block auto-merge, when you must decide whether a PR is safe to arm yet, or to handle the race where review comments land after a PR is already queued.

0 installs
▌
Skill Creator · skyfox675 bundle
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

0 installs
▌
Doc Coauthoring · skyfox675 bundle
Guide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks.

0 installs
▌
Gh Issue Filing · skyfox675 bundle
Write and triage GitHub issues that AI sub-agents can implement without asking a single follow-up question — the six-section groomed-issue format (verified file:line evidence, literal acceptance criteria, traceability hint), the filing-time label taxonomy, duplicate and cross-reference etiquette between concurrent operators, follow-up filing instead of scope creep, and the recon protocol that classifies stub issues into ready-to-dispatch / needs-spec-input / blocked via a sandboxed cheap-model decision tree with a machine-parseable output contract. Use whenever you are about to file, draft, groom, split, enrich, or triage a GitHub issue or a batch of them — when an operator reports a bug or pastes an error, when a feature request needs to become trackable work, when you discover adjacent work mid-task that must become a follow-up issue rather than scope creep, or when an issue queue needs triage before dispatch — even if the user just says "file this", "make a ticket", or "write that up".

0 installs
▌
Gh Issue Labels · skyfox675 bundle
Defines and operates the label-based control plane for multi-agent GitHub orchestration — agent-model:* (per-issue model override), agent-effort:* (effort directive carried in the dispatch prompt, not an API param), agent-claimed (active lock marker), do-not-dispatch (operator hold), do-not-rebase (auto-rebase opt-out), and ready-to-dispatch (advisory grooming signal). Use whenever you are selecting an issue to dispatch, resolving which model or effort tier a sub-agent should run at, deciding whether an issue may be claimed at all, applying or honoring an operator hold, breaking an auto-rebase starvation loop on a PR, or bootstrapping these labels in a new repo (scripts/bootstrap-labels.sh) — even if the user never says "labels"; every dispatch or claim decision must consult this taxonomy first.

0 installs
▌
Grooming Issues · skyfox675 bundle
Interactively groom an issue that is not yet ready for development — engage the engineering/project lead with questions to assert the intent, scope, and direction of the ask until ≥90% certain the written story reflects what they want, then fill the issue body to the six-section groomed anatomy. Captures the WHAT and WHY (product/spec intent), upstream of the technical-recon skill which decides the HOW and the LoE. Hard rule: never modify acceptance criteria that have already been written unless the operator explicitly asks — existing AC is an approved contract. Use whenever a lead hands you one or more raw, vague, or half-written issues to 'groom', 'flesh out', 'clarify', 'tighten the story', or 'get ready for tech review' before any technical recon or dispatch; asks via an interactive question tool when one is available, otherwise via emoji-answerable issue comments, and resumes on rerun once the lead answers.

0 installs
▌
Technical Recon · skyfox675 bundle
Technical recon of an already-groomed, stakeholder-accepted issue that the dev team has not yet vetted or sized — dispatch a read-only sub-agent (one per issue, in parallel for a batch) to trace the ask into the codebase, produce a verified implementation approach, assert a level-of-effort (LoE) estimate with confidence, surface risks/unknowns/dependencies, recommend a split when the work is too big for one PR, and pin the eventual dispatch model/effort/lane — then post the findings to the issue and move it to a dev-vetted state. Use whenever an engineering or project lead hands you one or more issue numbers/keys to 'recon', 'scope', 'size', 'estimate', 'tech-review', or 'assess implementation' before any code is written; also use when deciding whether a groomed issue is actually ready for a dev dispatch. This is the deeper dev-side pass, distinct from the cheap intake-triage recon in the issue-filing skill.

0 installs
▌
CI Flake Hunting · skyfox675 bundle
The authoritative method for hunting, root-causing, and fixing CI flakes so the merge queue stops ejecting correct PRs and churn drops. A flake is a test that fails then passes on retry with NO code change explaining it — every red speculative merge-queue build ejects a correct PR and disarms its auto-merge, so a live flake taxes every PR behind it. Encodes the loop — MINE CI history for real flakes → REPRODUCE harsher-than-CI → ROOT-CAUSE the real mechanism (hydration/render race, settle race, mock-order, loading-state race, click-through fragility) → FIX FORWARD (test or component, never mask) → VERIFY by re-running many times → UNSTICK the PRs the flake stranded → REPEAT. And how to tell a true code flake from transient INFRA (CDN/mirror outage, expired token mid-suite, rate-limit/WAF, external-service cap, cloud state-lock, CI-provider billing/outage, upstream bump) which must NOT be code-fixed. Use whenever chasing CI instability, "deflake this spec", "CI keeps ejecting my PR", "is this a flake or real",

0 installs
▌
CI Speed Hunting · skyfox675 bundle
The authoritative method for hunting, diagnosing, and fixing CI SLOWNESS so the merge queue drains faster and throughput rises — the target is wall-clock latency, NOT correctness (a job that is GREEN but slow). Encodes the loop — MINE CI timing for the real long poles → DIAGNOSE the bottleneck class (under-sharded matrix, cold cache, missing/too-broad path-filter tier, serial needs-chain, redundant rebuild) → PROVE the fix keeps identical coverage → FIX FORWARD → VERIFY the before/after delta at unchanged coverage → REPEAT. And the hard line — speed comes from doing the same work faster or skipping PROVABLY irrelevant work, NEVER from deleting tests, narrowing greps, dropping shards, or removing a gate. Use whenever chasing CI latency, "CI is too slow", "speed up the checks", "why does the merge queue take so long", re-sharding an E2E matrix, adding or fixing a cache, tuning changed-paths tiering, or running the ci-speed-hunter lane. Distinct from ci-flake-hunting (which owns failures and flakes — speed never

0 installs
▌
Gh Issue Locking · skyfox675 bundle
Claim/lock/release protocol for GitHub issues worked by AI agents, with multi-operator coordination etiquette. Use whenever an orchestrator is about to claim, dispatch, or start work on an issue (lock BEFORE dispatch, always); whenever a PR linked to a claimed issue merges or closes (release the lock and explicitly close the issue — GitHub auto-close does not fire on non-default integration branches); whenever you find an issue carrying a lock label or lock comment and must decide active-vs-stale; and during any batch flow that claims multiple issues in one pass — even if the user only says "pick up the next issue", "work #N", "fix", "release the lock", or "why is this issue still open". Also covers bundled-issue locks, stale-lock reclaim, the multi-claim race re-check, and what to do when another operator already holds the lock.

0 installs
▌
Spelunking Specs · skyfox675 bundle
Deep-dive an unknown or evolving codebase and distil it into living spec documents — spawn agents (one per discovered domain) to ingest the code, infer the conventions, standards, assumptions, and invariants, verify them against real file:line evidence, and write them as numerically-referenced domain specs (1, 1.2, 1.2.3 = domain.subdomain.instance) that live in the repo beside the code and are refined forward over time as more code and issues appear. Each spec captures discovered reality AND the standard it implies, flagging where the code drifts from its own pattern. For convention-, config-, or runtime-wired stacks where relationships are machine-names, hooks, events, and DI bindings rather than direct call sites, recon uses the framework's own introspection tooling and live runtime observation to recover the wiring that static reading misses — automating the manual 'spelunking' people otherwise do by hand. Use whenever you need to map, understand, or onboard to an application you don't fully know, establi

0 installs
▌
Browser Diagnosis · skyfox675 bundle
Reproduce and root-cause a front-end bug in a real browser using whichever browser MCP is connected — Chrome (claude-in-chrome), Playwright, Cypress, or Chrome DevTools — not just one. Drive the page as the signed-in user, watch the network and console, capture the failing request/response, shim/mock to isolate the fault, and record a repro, then hand the evidence to the issue being filed. Use whenever a symptom needs live-browser evidence (the toast says failed but it saved, a column shows zero, a click 404s) and from the /…-use-browser command family (and the engine-pinned /…-use-chrome, /…-use-playwright, /…-use-cypress variants), the /…-use-aws-browser commands, and the diagnostician agent. Picks the available MCP automatically and degrades gracefully when none is connected.

0 installs
▌
Jira Issue Fields · skyfox675 bundle
Defines and operates the field- and label-based control plane for multi-agent Jira orchestration — issue type (bug/story/task selection), the native Priority field (P0→Highest … P3→Lowest), agent-model-* labels (per-issue model override), agent-effort-* labels (effort directive carried in the dispatch prompt, not an API param), agent-claimed (active lock marker), do-not-dispatch (operator hold), do-not-rebase (auto-rebase opt-out, lives on the GitHub PR not the Jira issue), and ready-to-dispatch (advisory grooming signal). Use whenever you are selecting an issue to dispatch, resolving which model or effort tier a sub-agent should run at, deciding whether an issue may be claimed at all, applying or honoring an operator hold, breaking an auto-rebase starvation loop on a PR, or setting up these fields/labels in a new Jira project — even if the user never says "labels"; every dispatch or claim decision must consult this taxonomy first.

0 installs
▌
Jira Issue Filing · skyfox675 bundle
File and triage Jira issues that AI sub-agents can implement without asking a single follow-up question — the six-section groomed-issue format (verified file:line evidence, literal acceptance criteria, traceability hint) in the issue description, the filing-time issue-type/Priority/label taxonomy, duplicate and cross-reference etiquette between concurrent operators, follow-up filing instead of scope creep, and the recon protocol that classifies stub issues into ready-to-dispatch / needs-spec-input / blocked via a sandboxed cheap-model decision tree with a machine-parseable output contract. Use whenever you are about to file, draft, groom, split, enrich, or triage a Jira issue or a batch of them — when an operator reports a bug or pastes an error, when a feature request needs to become trackable work, when you discover adjacent work mid-task that must become a follow-up issue rather than scope creep, or when an issue queue needs triage before dispatch — even if the user just says "file this", "make a ticket",

0 installs
▌
Jira Issue Locking · skyfox675 bundle
Claim/lock/release protocol for Jira issues worked by AI agents, with multi-operator coordination etiquette. Use whenever an orchestrator is about to claim, dispatch, or start work on a Jira issue (lock BEFORE dispatch, always); whenever a PR linked to a claimed issue merges or closes (release the lock and explicitly transition the issue to Done — do not rely on Jira's PR-merge auto-transition); whenever you find an issue carrying a lock label or lock-marker comment and must decide active-vs-stale; and during any batch flow that claims multiple issues in one pass — even if the user only says "pick up the next issue", "work PROJ-123", "fix", "release the lock", or "why is this issue still open". Also covers bundled-issue locks, stale-lock reclaim, the multi-claim race re-check, and what to do when another operator already holds the lock.

0 installs
▌
Orchestrating Slots · skyfox675 bundle
Run an N-slot, event-driven multi-agent orchestration loop over a repo — survey PRs and issues, fill concurrency slots with sub-agent dispatches P0-first, react to merge/CI/review events, apply backpressure when the queue jams, and keep all state in labels/fields, assignees, timestamps, and a sentinel pause issue so any session can resume cold. Use whenever the operator asks you to orchestrate, "fill your slots", "work the queue", "keep N agents going", run rounds or a loop, monitor/babysit the merge queue, or opens chat with bare "fix" and no specific issue — even if they never say the word "orchestrate". Also invoke when deciding whether to dispatch another agent, whether the queue is genuinely backed up, or how to interpret a PR's mergeStateStatus.

0 installs
▌
Driving Prs To Merge · skyfox675 bundle
Owns the entire post-open PR lifecycle in a multi-agent GitHub repo — enabling and re-arming auto-merge, reading mergeStateStatus correctly, triaging CI failures from per-job logs, classifying transient vs real failures, resolving bot and human review threads (including the async-bot orphan trap), rescuing DIRTY/conflicted PRs, operating merge queues, handling pre-existing red CI, and dispatching fresh-context peer review. Use whenever a PR has been opened and must reach merge, a PR is stuck (green-but-not-merging, BLOCKED, DIRTY, booted from a merge queue), CI on a PR is red, auto-merge "mysteriously dropped", review threads need handling, or any agent encounters failing CI — even if the user only says "why isn't my PR merging", "CI is red", "fix this PR", or "the queue looks broken".

0 installs
▌
Dispatching Subagents · skyfox675 bundle
Turn a groomed tracker issue into a running implementation sub-agent and a merged PR — issue selection (priority, file-disjointness, bundling), model/effort resolution from chat > labels > defaults, the eight-section dispatch brief (worktree, toolchain prefix, pre-push hygiene, scope, git discipline, PR procedure, return contract), parallel dispatch under a concurrency cap, stalled-agent worktree salvage, and verify-then-trust completion checks. Use whenever you are about to spawn a sub-agent (Agent/Task tool call) to implement, fix, refactor, test, or otherwise write code from a tracker issue — even if the user only says "work the next issues", "fix these bugs", "run the loop", or "send an agent at #N" without saying "dispatch"; also use when a dispatched agent stalls, is watchdog-killed, returns status:failed, or reports completion and you must decide whether to salvage its worktree, re-dispatch, or verify its PR.

0 installs
▌
Fast Forwarding Branches · skyfox675 bundle
Keep the primary checkout's integration branch current with a safe, interval-driven `git pull --ff-only` — and alert, never improvise, when the fast-forward is refused. Because agents work in worktrees cut from the primary, a stale primary silently hands every new worktree a stale base (stale hooks, stale scripts, stale skills). A refused fast-forward ("Your local changes to the following files would be overwritten", "Not possible to fast-forward") means someone left work in the primary or the branch diverged — that is an anomaly to report, never to stash, reset, rebase, or merge away. Use whenever running the branch-ff lane, keeping a local checkout in sync on a loop, or deciding what to do when a pull is blocked by local changes or divergence.

0 installs

30 published skills