agustinusnathaniel/maestria

v0.1.5MIT

Portable Maestria engineering methodology delivered as Agent Skills for compatible clients.

@maestria/agent-plugin

#312 998a9dd Thanks @agustinusnathaniel! - Clarify when Sonar research is complete: each requested question needs an evidence-backed answer or a specific unresolved gap. Blitz now reuses valid context and checks, switches routes when investigation or design is needed, and continues through verification, review, and delivery.

#304 93e4b3c Thanks @agustinusnathaniel! - Clarify rendered verification and visual delivery evidence, including documentation sites, CLI output, headless capture, and unavailable uploads. Keep required artifacts and verification gaps in handoffs, reconcile accepted requirements before completion, and apply independent review to meaningful direct implementation.

#278 c770954 Thanks @agustinusnathaniel! - Clarify installation paths, prerequisites, and support boundaries in package documentation. Separate contributor validation from user setup and remove stale pre-release wording.
#281 5c5729b Thanks @agustinusnathaniel! - Fix stale skill references so prescribed skills resolve at load time: planner skill guidance no longer names the removed to-issues and to-prd skills, and the duplicated Kimi orchestrator skill prescription (which restated canonical skill governance and drifted stale) is removed so sync.config.ts only adapts toward the plugin runtime. Clarify in the orchestrator's Role-Based Pipeline that @diagnose analyzes the bug, applies the minimal fix, and verifies the repair instead of grouping it with analyze-only thinkers. Correct the canonical agent-directives README index to count 8 pipeline agents (orchestrator + 7 specialists) and list orchestrator.md.
#280 7c038cd Thanks @agustinusnathaniel! - Make agent investigation, verification, skill loading, and output structure proportional to the task. Preserve complete assignment ownership and delivery, allow necessary in-scope regression coverage, and keep downstream instructions independent of Maestria repository paths.

#275 56351de Thanks @agustinusnathaniel! - Refine specialist engineering judgment around trust-boundary normalization, feature-local seams, shared-interface compatibility, executable sources of truth, staged migrations, durable diagnostic knowledge, and operational documentation.

#269 ef07bf7 Thanks @agustinusnathaniel! - Add conditional visual-evidence guidance for PR delivery: confirm the project targets GitHub with authenticated gh --attach support and an available capture tool before attaching a screenshot or short video for visual or behavioral changes. Vision is optional verification only, not a precondition.
#269 ef07bf7 Thanks @agustinusnathaniel! - Add a global testing-judgment rule (test artifacts are opt-in, behavior over implementation shape, mocks only at external seams) and align reviewer and diagnose guidance with it.

#265 4dcbf04 Thanks @agustinusnathaniel! - Add a first-class Agent Plugins v1 package that exposes Maestria's methodology as a portable, skills-only plugin, plus CLI commands to validate and stage portable packages.