Development

Write a reply or an independent opinion into a team-comms thread. Use when the user wants to answer a question from another member's agent, add a view to a discussion, or respond to something surfaced as unread.

Start a new discussion thread in the team's comms space so several members' agents can weigh in. Use when the user wants to ask the team something, raise a design question, or get other agents' opinions on a decision.

Declare that this working directory is now taking part in a team-comms thread, so future sessions here surface its unread posts - or stop, so they go quiet again. Use when the user says they are working in a thread, wants notifications for one, or wants them to stop.

Record the conclusion of a team-comms discussion and close the thread. Use when a discussion has reached an answer, when the user says what was decided, or when a thread is finished and should stop appearing as live.

Restore a team-comms thread's discussion into this session - the cached digest plus whatever arrived since - and save an updated digest. Use when starting work in a thread whose history this session does not have, after focusing one, or when the user asks what has been said.

Connect this machine and project to a team-comms discussion space on a shared folder - create .claude/team-comms.json, verify the drive is really readable, and register this agent. Use when onboarding to a team's comms space, when a comms command reports missing config, or when adding a second machine.

Keep memory small enough to stay useful - report which caps are over, propose merges, split bloated notes, archive what is finished, and connect orphans. Use when memory-doctor or the session brief reports a cap, when memory feels noisy, or on a periodic review.

Promote what a session learned from the short-term record into durable memory - review recent episodes and conversations, keep what will matter next time, and write it as typed notes. Use at the end of a session, when the user asks to reflect/consolidate/remember what was learned, or when memory-doctor reports the episodes cap.

Check whether long-term memory is actually working on this machine - setup, engine copy, model cache, vault, database, and whether capture is still recording. Use when memory looks stale or empty, when sessions stop being remembered, before reporting a memory bug, or when the user asks to check/diagnose memory.

Design a change or feature with the user and write the design set — proposal, per-capability specs, design (with open questions) and UML diagrams (use case, ER, sequence, file change map) — into a folder they choose. Settles the design first by interviewing the user one round at a time. Use when the user wants to design a system, feature or change before building it, asks for a design doc / spec / proposal, or wants requirements turned into design documents.

Review Archify issues, PRs, or code through value, cost, and impact to support evidence-based maintenance decisions. Use for issue triage, change reviews, and code quality assessments.

Automated, compiler-accurate AST refactoring and migration engine for Kotlin Multiplatform and Gradle projects using OpenRewrite. Use when performing repo-wide dependency upgrades, version catalog (libs.versions.toml) updates, Gradle deprecation remediation, or automated framework migrations (e.g. Ktor 2 to 3, Coroutines, or Java to Kotlin brownfield upgrades) across dozens of multiplatform modules in a single step.

Govern GitHub issue creation, sub-issue hierarchy, and status updates for KMP repositories. Use when decomposing work into epics and sub-issues, preventing ticket spam or comment flooding, drafting bug reports and technical tasks, or transporting markdown payloads safely through the GitHub CLI without escaping corruption.

Enforce Definition of Ready (DoR), Definition of Done (DoD), and quality verification gates for Kotlin Multiplatform engineering. Use before picking up a task or feature to verify prerequisites, during implementation to validate UI and performance constraints, and before opening or merging Pull Requests to guarantee release-quality standards.

Move a WordPress site's media to S3 with the S3 Uploads plugin — installs the plugin with its vendor tree, writes s3-config.php and the endpoint mu-plugin, migrates the existing library with a transfer that verifies itself, and reverses the whole thing. Use when the user wants uploads on S3, object storage, MinIO or any S3-compatible server, mentions S3 Uploads, offloading media, a media CDN on CloudFront, or asks to sync wp-content/uploads with a bucket. Run through /wp-s3 and /wp-s3-media.

Local SEO audit reference — business-type and vertical detection, NAP consistency sources, LocalBusiness subtype selection, location-page quality gates and citation tiers, with the WordPress option and meta keys each check reads

Review code for architectural concerns only — boundary violations, wrong-layer logic, undocumented dependencies, and implicit decisions that should become an ADR or RFC. Works on a PR/branch/diff, or on standing code (a package, extension, or subsystem that already exists). Use when asked to "architecture review" something, to check whether code respects Silo's boundaries, before merging something cross-cutting, or when auditing an existing area for drift. Not a code-quality or style review.

Bootstrap BuildOS governance where lifecycle hooks are unavailable. Use when establishing a project governance baseline; ordinary tasks use project entrypoints.

Bootstrap BuildOS governance where lifecycle hooks are unavailable. Use when establishing a project governance baseline; ordinary tasks use project entrypoints.

Users versus admins in an ibl.ai app — the User/Admin view switch, the isTenantAdmin() rule, an admin area (users and invitations, roles and policies, analytics, billing, memory, organization) that only org admins see, and where each admin surface comes from in the SDK. Use when the user mentions admin, administrators, user management, roles, permissions, invite users, admin dashboard, "who can", or wants some pages hidden from regular members. For agent-level sharing see /iblai-vibe-agent-ac...

Manage an ibl.ai organization's per-user feature configuration and app/onboarding state via the Data Manager (DM) API — read and write a user's feature config (single feature or bulk inline object), bulk-update a feature's values across an entire platform, list the apps a user can access, mark onboarding complete, activate an app free trial, and (when enabled) run an admin platform-provisioning config. Use when toggling feature flags/values for a user or a whole org, reading which apps a user...

Start here for anything on the ibl.ai platform — the map from what you want (a new app; users and admins; custom user or organization data; agents; memory; analytics; charging users; iOS, Android, macOS, Windows; operating the platform headlessly from a terminal or CI; anything with no component) to the skill that does it. Use when the user mentions ibl.ai, iblai, or vibe, points an agent at github.com/iblai/vibe, asks where a capability lives, which skill to use, whether a skill has a UI, or...

The first conversation for anything on ibl.ai — four questions that decide the whole shape of the work before any file is touched (new project or existing codebase; single-organization app, multi-organization app, or headless server-to-server; who signs in — members, the public, or nobody; which of users, memories, agents, and organizations the app is about), recorded in iblai.env and the project CLAUDE.md, then routed to the right skills. Use when the user says "start", "I want to build", "n...

Map of the ibl.ai open-source ecosystem — the repo family and the shared @iblai tooling, all built on the api.iblai.app backend. iblai/vibe (this repo) holds both skill families — iblai-vibe-* to build apps with visual components and iblai-api-* to operate the platform headlessly (the former iblai/api); iblai/os and iblai/lms are complete forkable sample apps; iblai/iblai-infra-cli self-hosts the backend. Read this to orient: which repo to reach for when operating vs. building vs. forking vs....