Codeseys Labs
Official@Codeseys-Labs
Codeseys Labs publishes skills for orchestrating multi-agent software development lifecycles, stacked pull requests, decision records, threat modeling, and repository quality gates.
Agent Skills by Codeseys Labs
Showing 13 vetted skills indexed across 1 GitHub repositories.
Codeseys-Labsagentic-sdlc
Orchestrates multi-agent software delivery waves using Seeds queues, Git worktrees, and provider-native delegation.
Codeseys-Labscodex-research-os
Installs and operates a repo-native autonomous research organization with specialist agents, claim ledgers, and review gates.
Codeseys-Labsrepo-toolchain-gates
Configures pinned mise toolchains, lefthook git hooks, and betterleaks secrets gates for repositories.
Codeseys-Labsstacked-prs-gh-cli
Manages dependent GitHub pull request stacks using plain gh CLI and git rebase.
Codeseys-Labsexternal-skill-libraries
Install third-party skill libraries through their own front doors with collision prechecks.
Codeseys-Labsdispatching-exact-ocx-models
Injects and verifies exact OCX model routes when dispatching Agent or Workflow workers.
Codeseys-Labsstacked-prs
Manages stacked pull request chains with safe rebasing, retargeting, and leased branch deletion.
Codeseys-Labsadr-lifecycle
Author, review, number, index, and supersede Architecture Decision Records.
Codeseys-Labssdlc-threat-model
Applies STRIDE lenses to one bound trust boundary and emits evidence-graded findings reports.
Codeseys-Labsmodel-tier-rightsizing
Routes agent dispatches to certified model tiers by wrong-output blast radius and verification strength.
Codeseys-Labscmux-event-bus-messaging
Implements pub/sub messaging between agent workers over the cmux event bus.
Codeseys-Labsreviewing-overengineering
Audits plans and diffs for accidental complexity while preserving safety-critical properties.
Codeseys-Labschange-writing
Drafts commit, pull-request, and squash messages from verified repository evidence.
Frequently Asked Questions About Codeseys Labs
FAQPage SchemaWhat tasks can I accomplish with Codeseys Labs skills?▼
You can run planned multi-agent SDLC waves with worktree isolation, manage stacked pull requests via gh, enforce mise/lefthook/betterleaks quality gates, author and supersede ADRs, run STRIDE threat models on scoped diffs, audit overengineering, and compose policy-compliant commit and PR messages.
Who are these skills designed for?▼
They target engineers and technical leads driving agent-assisted software delivery: implementers, reviewers, integrators, and critics working in SDLC-activated repositories who need disciplined orchestration, review gates, decision records, and threat analysis rather than improvised multi-agent sessions.
How do the agentic-sdlc and gate skills run in practice?▼
agentic-sdlc fires when a session opens an SDLC-activated repository with open seeds, using the host's native agent and subagent capabilities with no cmux or tmux required. repo-toolchain-gates makes `mise run check` the single gate, with lefthook hooks and betterleaks secrets scanning wired into CI.
What are the prerequisites and dependencies?▼
The baseline requires only the host's native agent capabilities. repo-toolchain-gates expects mise, lefthook, and betterleaks; stacked-prs-gh-cli expects plain gh and git. cmux-event-bus-messaging applies only when cmux is already active or explicitly requested, and external skill libraries install strictly opt-in.
Do these skills take destructive actions on my repository?▼
No. change-writing emits proposed message text only and never stages, commits, pushes, or merges. sdlc-threat-model is advisory and files no seeds itself, and accepted ADRs never authorize pushes, merges, or deployments. External library installs are dry-run by default with an explicit --yes flag.