Enterprise shared library of GitHub Copilot customization assets — agents, skills, instructions, and prompts.
Agent Claude Code
Coordinate multiple concurrent Copilot worktree sessions executing independent issues in parallel, with merge serialization and conflict prevention. USE FOR: parallel feature sprints, multi-worktree execution, serialized merge pacing, merge-conflict prevention. DO NOT USE FOR: single-session work, direct code…
Agent Claude Code
Dependency-aware unblock lane coordinator. USE FOR: reordering open PRs/issues by dependency to unblock broken builds fast, cherry-picking blocker fixes from issue and PR queues into a focused unblock group, and returning work to normal queue order after verification. DO NOT USE FOR: sprint capacity planning…
Agent Claude Code
Use when enforcing release freeze windows, routing exception requests, and coordinating branch access during a freeze. USE FOR: evaluate freeze exceptions, block unauthorized merge attempts, and publish go/no-go decisions for frozen branches. DO NOT USE FOR: merging changes, bypassing approvals, or resolving code…
Agent Claude Code
Assesses release readiness, change impacts, blast radius, rollback planning, and safe deployment strategies with canary deployments, feature flags, and changelog generation. USE FOR: assess release blast radius, plan rollback strategy, recommend deployment approach. DO NOT USE FOR: executing deployments, live incident…
Agent Claude Code
Automated versioned release workflow. Reads merged PRs since the last release, bumps version.json, writes CHANGELOG entry, creates git tag, and publishes GitHub release. USE FOR: bump semver and publish GitHub release, generate changelog from merged PRs, create git release tag. DO NOT USE FOR: sprint planning…
Agent Claude Code
Use when facilitating release readiness ceremonies and making explicit go/no-go decisions. USE FOR: collect release evidence across quality, operations, and rollback readiness; identify unresolved launch risks; enforce gate criteria; and publish decision records with owners and due dates. DO NOT USE FOR: implementing…
Agent Claude Code
End-of-sprint retrospective agent. Reviews closed issues and merged PRs, produces Went Well / Improve / Action Items summary, and files improvement issues. USE FOR: run end-of-sprint retrospective, generate sprint improvement summary, file BaseCoat improvement issues. DO NOT USE FOR: planning next sprint, velocity…
Agent Claude Code
BaseCoat consumer rollout and update specialist. USE FOR: onboarding or updating a BaseCoat consumer, installing immutable BaseCoat releases, configuring downstream update notifications, creating guarded upgrade PR policy, and validating consumer sync provenance. DO NOT USE FOR: editing BaseCoat framework internals…
Agent Claude Code
Cleans GitHub Actions workflow run history safely by deleting stale, completed runs according to retention policy while preserving evidence for active incidents and compliance windows. USE FOR: pruning old completed workflow runs, reducing run-history noise, enforcing run-retention policy, and producing a cleanup…
Agent Claude Code
Automated CI failure analysis, log parsing, and pipeline remediation with retry strategies, flaky test detection, dependency resolution, and cache invalidation. USE FOR: auto-remediate CI failures, quarantine flaky tests, resolve build dependency and cache errors. DO NOT USE FOR: designing CI pipeline architecture…
Agent Claude Code
Persistent fleet delivery loop coordinator for ship-it sessions. USE FOR: bounded wave execution across tasks/PRs/checks, cycle status summaries with stop conditions, transient retry handling with escalation, and dry-run loop planning. DO NOT USE FOR: bypassing required checks, direct production deployment without…
Agent Claude Code
Intent-to-production orchestrator converting ship-it, spec-2-prod, and onboarding-conductor goals into governed execution loops with tracked PR, validation, release, and learning artifacts. USE FOR: goal-driven spec-to-prod orchestration, onboarding conductor discover/plan/apply/validate loops, build-break recovery…
Agent Claude Code
Task scope validator for sub-agent dispatch. Analyzes task prompts to detect overscope, ambiguity, and risk before forwarding to explore, task, or general-purpose agents. USE FOR: validate task prompts pre-dispatch, classify tasks as automatable/gather-only/defer, identify scope refinement needs. DO NOT USE FOR…
Agent Claude Code
Data architecture design specialist. USE FOR: designing data models and schemas, planning data warehouse architecture, optimizing query designs. DO NOT USE FOR: ETL operations, data pipeline troubleshooting.
Agent Claude Code
Distributed data integrity patterns — eventual consistency strategies, conflict resolution, ACID compliance, backup verification, and data recovery procedures. USE FOR: distributed data integrity patterns, consistency strategies, conflict resolution, backup verification, and recovery planning. DO NOT USE FOR…
Agent Claude Code
Data persistence and storage optimization. USE FOR: optimizing database configurations, designing backup strategies, managing data retention. DO NOT USE FOR: application development, real-time operations.
Agent Claude Code
Database migration and modernization specialist. USE FOR: planning database migrations, designing migration strategies, validating data integrity. DO NOT USE FOR: operational database management, routine backups.
Agent Claude Code
Data operations and pipeline management. USE FOR: managing data pipelines, monitoring data quality, optimizing data flow. DO NOT USE FOR: data analysis, business intelligence.
Agent Claude Code
Code review and quality gate specialist. USE FOR: reviewing code changes, enforcing quality standards, suggesting improvements. DO NOT USE FOR: writing code, direct fixes.
Agent Claude Code
E2E Test Strategy Agent for end-to-end testing orchestration, critical path identification, flakiness prevention, and cross-browser coverage. Covers Playwright, Cypress, Selenium patterns and integration with CI/CD pipelines. USE FOR: design E2E suite, map critical user paths, fix flaky tests. DO NOT USE FOR: unit…
Agent Claude Code
USE FOR: validating a BaseCoat guidance draft before committing, checking lint rules on agent/skill/instruction/prompt files, auditing frontmatter schema compliance, verifying BaseCoat conventions, returning pass/fail verdict with actionable fixes. DO NOT USE FOR: writing new guidance assets, scanning repos for…
Agent Claude Code
Produces structured manual testing strategies for features and risk inventories. USE FOR: decision rubrics, exploratory charters, regression checklists, defect templates, and automation backlogs. DO NOT USE FOR: automated test implementation, production incident command, or replacing acceptance criteria.
Agent Claude Code
Security penetration testing specialist. USE FOR: designing penetration tests, identifying security vulnerabilities, generating security reports. DO NOT USE FOR: fixing vulnerabilities, incident response.
Agent Claude Code
Code-level resilience pattern review — circuit breakers, timeouts, bulkhead isolation, graceful degradation, retry logic, and load shedding implementation. USE FOR: review circuit breaker and retry patterns in code, audit timeout hierarchy, validate graceful degradation. DO NOT USE FOR: live incident response…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: