swiftpostlabs
OrganizationShareable skills and tools for AI agents
Categories
Indexed Skills (46)
ref-sp-dev-playwright-cli
Automate browser interactions, test web pages, and work with Playwright tests via the playwright-cli terminal tool. Use when: driving a real browser from the command line (open/goto/click/fill/snapshot), debugging a page's console/network/DOM, running or debugging Playwright tests, recording video or traces of a session, or asking the user for live UI/design feedback.
tool-sp-commit
Inspect edited files, group them into logical commits, and create focused commits. Use when: the user asks to commit changes, split work into focused commits, or decide how the current diff should be grouped before committing.
tool-sp-create-skill
Create a new skill using the repo's skill-authoring standard. Use when: the user wants to add a new skill, scaffold a skill folder, turn repeated guidance into a skill, or run a guided wizard before writing skill files.
tool-sp-handle-agents-local-tasks
Read the local `.agents/tasks/` workspace — the `TODO.md` notes list and the tracked task folders under `new/`, `open/`, `closed/` — choose the next actionable item, and work it through with code changes, validation, folder moves, and status updates. Use when: the user asks to check `.agents/tasks/TODO.md`, continue remaining local tasks, or process the repo's local agent-task backlog.
tool-sp-maintain-agents-instructions
Review and update repository agent instruction files after code, workflow, or skill changes. Use when: AGENTS.md, GEMINI.md, or .claude/CLAUDE.md may be outdated, the skill catalog changed, or a multi-provider repo needs its instruction bridge refreshed.
tool-sp-maintain-skills
Review, consolidate, and update project skills after repo, workflow, or branch changes. Use when: skills may be outdated after code or tooling changes, guidance is duplicated or misplaced, or the repo's skill catalog needs a maintenance pass.
tool-sp-make-skill-shareable
Review an existing skill, decide whether it can be made shareable, and update its shareability metadata through a guided workflow. Use when: a skill lacks shareable-skills metadata, a user wants to export or symlink a skill, the current portability is unclear, or a repo-local skill might need to be split into a shared core.
ref-sp-agents-hooks
Author agent lifecycle hooks that run deterministic shell commands at session, prompt, tool, and stop events across Claude Code, GitHub Copilot CLI, VS Code, and Gemini CLI. Use when: creating or editing a hook, choosing a lifecycle event, writing a hook script that reads stdin JSON and returns an allow/deny/context decision, making a hook portable across agents, or debugging why a hook does not fire or block.
ref-sp-agents-instructions-authoring
Guidance for structuring and maintaining repository instruction files across major agent entry points such as Copilot, Gemini, and Claude. Use when: designing the repo's instruction system, choosing a source of truth, or updating AGENTS.md, GEMINI.md, and .claude/CLAUDE.md together.
ref-sp-agents-local-tasks
Reference guidance for using `.agents/tasks/` as a gitignored local workspace: a lightweight `TODO.md` notes list plus tracked task folders that move through `new/`, `open/`, and `closed/` lifecycle subfolders. Use when: reading or updating `.agents/tasks/TODO.md`, creating or moving a tracked task folder between `new/`, `open/`, and `closed/`, setting a task's frontmatter `status`, or checking whether local agent task notes still match the active work.
ref-sp-agents-mr-wolf-persona
The Mr. Wolf persona: the agent's voice, directness, and escalation stance — assess before acting, state your understanding before executing, correct wrong premises on sight (including the user's), stay blunt about problems and courteous to people, and report what is actually true rather than what lands well. Carries the canonical persona text that instruction files inline verbatim. Use when: starting or resuming a task, delivering unwelcome technical feedback or pushing back on a flawed premise, refreshing instruction files that must preserve the agent's voice, or adopting this working style in another repo.
ref-sp-agents-plugin-marketplaces
Publish a repo's skills as an agent plugin distributed through a plugin marketplace, installable from Claude Code, GitHub Copilot CLI, and VS Code: plugin.json and marketplace.json layout, the manifest detection order that lets one repo serve every client, source types, the skills-path and caching rules, how visibility tiers map onto public/private/local hosting, and how releases and updates actually reach users. Use when: packaging skills as a plugin, creating or hosting a marketplace.json catalog, deciding which skills may be published, targeting Copilot CLI or VS Code users, cutting or versioning a plugin release, or debugging why an installed plugin is missing skills or not updating.
ref-sp-agents-policy
Repository-specific guidance for the agents-policy feature, .agents/config.json policy section, and generated AI restriction outputs in this repo. Use when: working on the agents-policy feature, updating policy docs, or debugging Copilot, Claude Code, or Gemini policy generation here.
ref-sp-agents-security
Reference guidance for agent security policy, protected file access, exclusion sync, and multi-client enforcement. Use when: modifying a policy source file, updating the sync workflow, reviewing generated restriction files, or changing how agents are prevented from reading sensitive files.
ref-sp-agents-shareable-skills
Normative spec for how a repo names, describes, scopes, shares, and vendors skills: owner-prefix naming grammar, the domain registry, visibility tiers, dependency semantics, vendoring vs forking, and the sharing-spec validator. Use when: naming or renaming a skill, setting owner/domain/visibility/tags, recording hard vs soft skill dependencies, vendoring or forking a skill from another repo, or validating a skill for export.
ref-sp-agents-skills-authoring
Reference guidance for creating and maintaining project skills. Use when: designing new skills, updating existing skills, establishing skill standards, evaluating skill quality, organizing skill subfiles, or adapting copied skill content to the repo's actual stack.
ref-sp-agents-verification-discipline
Portable verification discipline that counters jumping to answers, sycophancy, and overconfidence: enumerate candidate approaches or root causes, route verification by confidence and stakes, prune on evidence, abstain explicitly when nothing can settle a claim. Use when: choosing between approaches or root causes, acting on an unverified claim or assumption, responding when the user challenges or contradicts a conclusion, deciding how much verification a risky or irreversible action needs, or calibrating stated confidence in answers, reviews, and reports.
ref-sp-baas-supabase
Portable Supabase guidance for CLI workflows, migrations, supabase-js CRUD usage, Edge Functions, and ORM boundaries. Use when: initializing Supabase, evolving schema, designing CRUD paths, writing Edge Functions, or deciding how ORMs fit with Supabase.
ref-sp-db-distributed
Portable distributed-database guidance for fragmentation, replica placement, local and global applications, parallel DBMS topologies, and disconnected synchronization. Use when: designing distributed or replicated databases, choosing site layout, planning mobile or intermittently connected data flows, or separating topology decisions from data-model decisions.
ref-sp-db-nosql
Portable NoSQL guidance for choosing document, key-value, graph, or wide-column models, handling CAP and eventual-consistency tradeoffs, and modeling denormalized access patterns deliberately. Use when: relational joins stop matching the workload, horizontal scale or flexible schema dominates, or a distributed data model is being designed.
ref-sp-db-operations
Portable guidance for running a database as a durable system of record: transaction boundaries and ACID, concurrency and isolation, indexing and storage, query optimization, backup and recovery, and safe schema migration. Use when: planning transactions or isolation levels, diagnosing lock contention or a slow query, choosing or pruning indexes, planning a migration or rollback, or reviewing whether backups actually restore.
ref-sp-db-schema-design
Portable guidance for designing a relational schema end to end: fact-finding and mission scope, conceptual ER/EER modeling, logical derivation and normalization, and physical design driven by the workload. Use when: starting a schema from requirements, turning business rules into entities, keys, and relationships, decomposing a wide table or reviewing functional dependencies and normal forms, deciding how far to normalize or when to denormalize, or reviewing whether a design skipped straight to implementation.
ref-sp-db-security
Portable database-security guidance for threats, access control, views, auditing, encryption, integrity, and secure recovery. Use when: protecting database-backed systems, designing authorization models, reviewing confidentiality, integrity, or availability risks, or securing backups, logs, and administrative access.
ref-sp-dev-coding-patterns
Portable coding defaults for readable, strongly-typed, well-tested code across languages and CLIs. Use when: choosing naming, typing, comments, branching structure, CLI ergonomics, or testing defaults in a new or existing codebase.
ref-sp-dev-docs-authoring
Portable README and documentation authoring guidance for top-level and feature-level docs. Use when: writing or restructuring a README, deciding whether user usage or developer setup should come first, adding concrete examples, or splitting long docs into focused follow-up files.
ref-sp-dev-git-commits
Reference guidance for grouping changes into focused commits and writing clear commit messages. Use when: deciding how to split changes into commits, writing a commit title or body, deciding whether a commit needs a long description, or documenting an automated command for reproducibility.
ref-sp-dev-github-actions-ci
Portable GitHub Actions CI guidance for workflow triggers, job design, permissions, matrix strategy, concurrency, caching, and workflow hardening. Use when: creating or reviewing .github/workflows/*.yml files, setting up CI or reusable workflows, or securing workflow tokens, actions, and runner choices.
ref-sp-dev-github-dependabot
Portable Dependabot guidance for dependabot.yml configuration, ecosystem selection, schedules, grouping, ignore or allow rules, private registries, and GitHub Actions updates. Use when: creating or reviewing .github/dependabot.yml, tuning update volume, or deciding how version and security updates should be grouped and scoped.
ref-sp-dev-projects-architecture
Portable architecture guidance for feature folders, boundaries, shared utilities, and separating product code from maintenance scripts. Use when: deciding where code should live, splitting features, or reviewing repository structure.
ref-sp-dev-repo-conventions
Repo-specific conventions for this Python project (agentic-tools): the feature-first src/agentic_tools layout, pyproject.toml configuration hub, Black + Pyright-strict + pytest tooling via Poe, typing rules, CLI and script placement, and translations. Use when: creating or moving features, tests, or CLI entrypoints; deciding which top-level or package folder a file belongs in; adjusting pyproject.toml, Poe tasks, or tool config; or explaining how this repo is wired.
ref-sp-dev-semantic-versioning
Portable semantic-versioning guidance for release numbering, prerelease handling, dependency range selection, and package.json dependency fields. Use when: choosing a version bump, reviewing semver compliance, setting npm version ranges, or deciding between dependencies, devDependencies, peerDependencies, optionalDependencies, and overrides.
ref-sp-js-deno
Portable Deno guidance for modern runtime usage, deno.json configuration, tsconfig and ESLint adoption, permissions, Node interop, and hybrid Deno or Node repositories. Use when: writing Deno code, configuring deno tasks, or adopting Deno into an existing TypeScript or Node repo.
ref-sp-js-javascript
Portable JavaScript guidance for scripts and browser code with JSDoc-based typing. Use when: writing plain JavaScript, adding JSDoc, or keeping JavaScript maintainable without TypeScript.
ref-sp-js-next-template
App-level guidance for building a full React and Next.js application. Use when: scaffolding or reviewing a whole React/Next app, choosing the baseline stack and package manager, deciding app-level structure, or comparing a full app against a simpler standalone browser tool.
ref-sp-js-next
Portable Next.js guidance for App Router structure, client and server boundaries, framework integrations, and Next-friendly library choices. Use when: creating or reviewing Next routes and layouts, deciding where 'use client' belongs, configuring Next.js, or choosing framework-specific integrations like next-intl.
ref-sp-js-react
Portable React guidance for components, hooks, client-side state, and React-friendly library choices. Use when: creating or reviewing React components, choosing React libraries, deciding where UI or async state should live, or refactoring a React feature that is getting hard to read.
ref-sp-js-typescript
Portable TypeScript guidance for strict typing, runtime boundaries, and maintainable scripts or apps. Use when: writing or reviewing TypeScript code, types, or configuration decisions.
ref-sp-js-userscript
Portable guidance for browser userscripts on Tampermonkey, Violentmonkey, and similar managers. Use when: writing or reviewing .user.js or .user.ts scripts, metadata blocks, permissions, or DOM automation.
ref-sp-js-web-standalone-template
App-level guidance for standalone HTML, CSS, and JavaScript tools that should run without framework or build-system overhead. Use when: creating or reviewing a whole browser-only app, deciding whether it can stay no-build, or choosing local assets and browser-loadable libraries.
ref-sp-py-commitizen
Python Commitizen release workflow guidance. Use when: configuring commitizen in pyproject.toml, choosing version providers, generating changelogs, validating conventional commits, or designing Commitizen-led release commands for Python or mixed Python/Node repos.
ref-sp-py-python
Portable Python guidance for typed application code, scripts, CLIs, and tests. Use when: writing or refactoring Python modules, designing Python feature folders, or deciding typing and testing patterns.
ref-sp-web-marketing
Measure web marketing honestly: why 'Direct' is a bucket of lost origins rather than a channel, how dark social destroys referrer data, when a conversion difference is noise rather than signal, and how to recover attribution you cannot track. Use when: reading a traffic, channel, or conversion report; asked which channel is working or where to spend budget; asked why Direct traffic is so high; interpreting a conversion-rate change, a campaign result, or a traffic spike; setting up UTM tags or self-reported attribution; analyzing analytics from Google Analytics, Wix, Shopify, or similar; or judging whether a marketing claim is supported by the data behind it.
ref-sp-web-seo-ai
Visibility in AI answers — Google AI Overviews and AI Mode, and assistants like ChatGPT, Claude, and Perplexity: how they retrieve and cite pages, how to control AI crawler access, how AI surfaces distort your Search Console numbers, and which GEO advice is defensible versus speculative. Use when: asked how to rank, appear, or get cited in AI search, AI Overviews, or chatbots; asked about GEO, AEO, or answer-engine optimization; deciding whether to allow or block AI crawlers (GPTBot, ClaudeBot, Claude-User, PerplexityBot, Google-Extended, CCBot), or whether to opt out of AI training; writing robots.txt rules for AI bots; or diagnosing falling clicks while impressions hold steady.
ref-sp-web-seo
Audit a website's SEO from what is actually observable — served and rendered HTML, robots.txt, sitemaps, canonicals, titles, structured data, Core Web Vitals — separate that from the performance data only Search Console can answer, and design before/after tests that mean something. Use when: asked to check, audit, or improve a site's SEO; diagnosing why pages are not indexed, not ranking, or why organic traffic dropped; reviewing robots.txt, sitemaps, canonicals, titles, or structured data; debugging why a crawler cannot see JavaScript-rendered content; SEO on Next.js, WordPress, or headless WordPress + Next.js (metadata, sitemaps, canonicals, static export); interpreting a Lighthouse or Core Web Vitals report; judging whether an SEO change actually worked; weighing an SEO claim from a leak, a vendor, or a blog; or assessing content quality against Google's rater framework.
ref-sp-web-social-media
Decide whether and how a site or brand should invest in social and video platforms (TikTok, YouTube, Instagram, LinkedIn, X) for organic discovery — separating what the platforms and independent research actually document from the large layer of agency folklore, and measuring it honestly inside walled-garden analytics. Use when: asked whether to be on TikTok/YouTube/Instagram for reach, how discovery or in-app search works on a platform, whether outbound links hurt reach, how younger audiences find things, how social content shows up in Google, or judging a social-strategy claim or agency deck.
ref-sp-web-wordpress
Build on WordPress correctly: the sanitize-input / check-capability / escape-output security model that prevents most WordPress vulnerabilities, hooks as the extension mechanism instead of editing core, safe database access with prepared statements, child themes, and the REST API for headless frontends. Use when: writing or reviewing a WordPress plugin or theme, handling form or request data in PHP, adding an admin action or AJAX/REST endpoint, querying the WordPress database, deciding how to extend WordPress without modifying core, auditing a WordPress site's security or update posture, or building a headless WordPress backend.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.