Extends skill installation to support 'templates/' and 'examples/'
subdirectories, in addition to the existing 'references/'.
The installer logic is updated to discover these new directories and
copy them to the target location during both initial install and
update operations.
This ensures that supplementary skill artifacts are available in the
installed environment, allowing features that rely on them to function
correctly.
feat(pm-team): add skill for Lerian Map platform integration
feat(installer): add support for pi-coding-agent to expand ecosystem
refactor(installer): remove Cursor and Cline support to simplify maintenance
docs: update skill counts to 93 and document new skills
refactor(skill): simplify name of lerian-map-management skill
style(skill): improve conciseness of dev-multi-tenant description
feat(hooks): add schema validator and enhance skill generation
refactor(*): align all components with the new frontmatter schema
refactor(commands): replace arguments object with simple argument-hint
refactor(agents): remove invalid version field from agent frontmatter
test(hooks): add unit tests for frontmatter validation and generation
Ran a full RED-GREEN-REFACTOR testing campaign (11 agent runs across
4 phases) that identified 5 gaps in the skill's enforcement coverage.
Additions:
- Anti-rationalization row: "Service has only one dependency"
- SaaS TLS Enforcement section: surface (probe) vs enforce (bootstrap)
- RabbitMQ TLS detection note: amqp091-go API limitation
- Next.js Self-Probe Lifecycle: instrumentation.ts register() pattern
- Runtime vs Startup table: complementary mechanisms clarification
Also restores /readyz convention (reverts prior /ready rename) to
align with K8s standard naming (/healthz, /livez, /readyz).
Generated-by: Claude
AI-Model: claude-opus-4-6
To improve clarity and align with more common industry standards, like
those in Kubernetes, the readiness probe endpoint is renamed from
/readyz to /ready throughout the skill documentation.
The response payload is also updated to use "healthy" and "unhealthy"
for status checks instead of "ready" and "not_ready". This change
creates better terminological consistency across health-related
endpoints.
All code examples, paths, and type definitions (e.g., ReadyResponse)
have been updated to reflect these changes.
Update the readiness probe example to use per-dependency timeouts.
This allows setting a stricter timeout for fast dependencies like caches
(1s) versus others (2s). The change improves probe accuracy by
preventing a slow cache from unnecessarily delaying the entire check.
The /health handler example is also updated to use a rich,
dependency-aware health check. This ensures the liveness probe
reflects the service's real-time ability to operate, not just a
static "ok" status, allowing for restarts if a dependency fails.
Adds ring:dev-readyz skill that ensures all Lerian services implement:
- /readyz endpoint with per-dependency status, latency, and TLS verification
- Startup self-probe that logs dependency checks and wires to /health
- Structured response contract consumable by Tenant Manager
Origin: Monetarie SaaS incident — product-console started without TLS
on MongoDB/DocumentDB, K8s liveness passed, client hit silent errors.
Covers Go (Fiber + lib-commons) and TypeScript/Next.js stacks.
Generates and audits AI-friendly entry point files for repositories:
- llms.txt following llmstxt.org spec (universal LLM entry point)
- CLAUDE.md for Claude Code / Anthropic agents
- AGENTS.md for vendor-neutral coding agents
Supports create, audit, and full (both) modes.
Includes content selection rules, audit checklists, and report format.
Generates and audits AI-friendly entry point files for repositories:
- llms.txt following llmstxt.org spec (universal LLM entry point)
- CLAUDE.md for Claude Code / Anthropic agents
- AGENTS.md for vendor-neutral coding agents
Supports create, audit, and full (both) modes.
Includes content selection rules, audit checklists, and report format.
Two misalignments found and fixed:
## 1. Pagination response field naming
Ring standards previously documented cursor fields as camelCase
(nextCursor, prevCursor, hasMore). The actual Midaz Pagination struct
uses snake_case json tags:
NextCursor string `json:"next_cursor,omitempty"`
PrevCursor string `json:"prev_cursor,omitempty"`
There is no HasMore/hasMore field in Midaz.
Changes:
- api-patterns.md: fix struct examples, JSON examples, detection
commands, tables, and anti-rationalization rows
- golang.md (opencode): pagination section was already correct
- PM skill files: update pagination field reference tables
## 2. BusinessError struct missing title field
Ring's BusinessError lacked the title field present in both
midaz/pkg/mmodel/error.go and lib-commons ErrorResponse:
type Error struct {
Code string `json:"code"`
Title string `json:"title"`
Message string `json:"message"`
EntityType string `json:"entityType,omitempty"`
}
Changes:
- platforms/opencode/standards/golang.md: add Title field to
BusinessError struct and to ErrInvalidInput/ErrNotFound examples
Files changed:
- dev-team/docs/standards/golang/api-patterns.md
- platforms/opencode/standards/golang.md
- pm-team/skills/pre-dev-api-design/SKILL.md
- pm-team/skills/shared-patterns/standards-discovery.md
Adds a mandatory discovery step during PRD creation to assess whether
the product/feature requires business dashboard endpoints.
Context: BTG Payments and Flowker pre-dev both missed dashboard
endpoints because nothing in the workflow prompted the question.
This resulted in gaps discovered late in the process.
The discovery step:
- Asks whether operators need aggregated visibility
- Documents dashboard requirements (metrics, consumers, refresh cadence)
- Provides guidance table for when dashboards make sense vs not
- Explicitly records the decision (yes/no) to prevent re-surfacing
Cascades into Feature Map (Gate 2) and API Design (Gate 4).
Suggested by: Garzão and Chacon in #the-ring-feedbacks
detect_python() now validates version >= 3.10 (needed for dataclass
slots=True). Falls back to a persistent .venv when only system Python
3.9 is available (e.g. macOS stock).
This commit overhauls the `lib-commons` skill documentation to reflect
all changes and new features introduced up to version v4.6.0. The goal
is to make the skill a current and comprehensive reference.
Key changes include:
- Add a new "Breaking Changes" section to provide a clear migration
path for developers upgrading from older v4.x versions.
- Document the event-driven tenant discovery model (v4.5.0+), which
replaces the legacy watcher system with a more efficient Redis
pub/sub approach.
- Add a detailed reference for the new `net/http/ratelimit` package,
including its configuration, tiers, and identity extractors.
- Explain the unified and variadic API for `tenant-manager` (v4.6.0),
which simplifies middleware setup and context retrieval for
multi-module services.
- Introduce a new section for the root `commons` package, covering the
`Launcher`, context helpers, string utilities, and env helpers.
- Expand the package catalog and "Which Package Do I Need?" guide to
include the 35+ packages now available.
