Menu
O SkillsMP coletou 10 skills de prompted365/sovereign-sidecar. Abra uma skill para revisar a origem e os detalhes.
Mostrando 10 de 10 skills coletadas.
Agentic citation intelligence and AI-search optimization for long-form publication. Trigger on /citation-intel, "is this citation ready", "what schema for", "optimize for citation", or any request to check discoverability / schema / token efficiency /…
Idioma do texto original: inglês
Closure inference and response-geometry disclosure — direction-agnostic, scope-aware detection of materially missing expressions around an active move. CENTROID: closure inference at the point of an active move IS: - post-landing closure check (detect…
Idioma do texto original: inglês
Adversarial falsification of an active move before it commits. Fire when an irreversible decision is pending — architecture choices, schema changes, API contracts, publication, deployment. CENTROID: adversarial falsification — what's the strongest case…
Idioma do texto original: inglês
Govern the fan-out to subagents — ensure the dispatch is a coherent spec, not an ad-hoc spray, before anything is spawned. CENTROID: orchestrator-as-delegator gate — govern the swarm/tranche spec before dispatch. IS: - the orchestration checkpoint: brief…
Idioma do texto original: inglês
Metabolize external context into the working frame without cargo-culting. Fire when external content enters the session — URLs, pasted docs, API references, competitor analysis, framework docs, tutorial code. CENTROID: metabolize external context — consume…
Idioma do texto original: inglês
Working-mode declaration and runtime contract — declare ENG/OPS × DIRECT/META posture at session start; enforce read-only semantics for META; surface scope-overreach warnings for DIRECT. CENTROID: session-mode contract that shapes which actions are allowed…
Idioma do texto original: inglês
Before mutation, detect whether the agent has enough local truth to act safely. CENTROID: repo-native pre-mutation gate — does the agent know enough to proceed? IS: - the practical pre-mutation readiness check (not philosophical, not post-move) - blast-radius…
Idioma do texto original: inglês
Theory of Mind (centroid-preserving) re-expression engine. Trigger on /tom, "run tom", "tom this for X", "rewrite for X", "compress for X", "reframe as X", "translate the centroid", or any request to re-express context-window content for a different audience,…
Idioma do texto original: inglês
Context consolidation — concatenate any file surface into a single LLM-consumable indexed markdown dump. Works with local dirs, glob patterns, git repos (ours or public), and conversational intent targets. CENTROID: file-surface to…
Idioma do texto original: inglês
Runtime tactical context hydration — staged discovery and bounded source-bearing hydration. Answers "how does the agent know where to look before it already knows where to look?" via filesystem shape, structural signals, and typed candidate baskets. CENTROID:…
Idioma do texto original: inglês