Menu
SkillsMP has collected 22 skills from samuelfaj/sam-skills. Open a skill to review its source and details.
Showing 22 of 22 collected skills.
Compile a host-safe quality-bar loop prompt and return it for the user to copy, edit, and paste. Never start the loop. Use when the user runs /sam-gauntlet-loop, says gauntlet loop, gauntlet this, or asks to loop until the work beats a real reference.
Finish a software goal completely with the smallest correct change: write checkable gates first, split independent units onto workers when the gate opens, verify every unit yourself, and never add a dependency. Use when the user runs /sam-goal, $sam-goal, or…
Delegate a bounded implementation task to the fixed Grok worker (grok-4.6) under workspace sandbox and headless execution. Use when a host agent or user needs Grok to implement, fix, or verify a scoped coding task; default effort high unless explicitly…
Claude-controller hybrid orchestration: Grok 4.6 workers for routine and deep slices (medium LIGHT / high STANDARD / xhigh DEEP), Claude opus high independent review, opus xhigh only for stall/multi-round capability escalation, and opus max for optional…
Codex-controller hybrid orchestration: Grok 4.6 workers for routine and deep slices (medium LIGHT / high STANDARD / xhigh DEEP), Codex gpt-5.6-sol medium independent review, and Sol high only for stall/multi-round capability escalation. Use when the user runs…
Coordinate complex work as a controller-only orchestrator using cost- and risk-aware capability routing, explicit task dependencies and ownership, skeptical proof verification, and an independent review gate. Use when the user asks for delegated execution,…
Conduct task study and emit a machine freeze plan (goal, thesis, steps, evidence, status) plus a required light-theme HTML pack for humans; assertive investigation first, council only on risk triggers. Use when the user runs /sam-plan, asks for an…
Consult Claude as a read-only advisor for a focused assumption, tradeoff, architecture question, security concern, difficult diagnosis, or high-risk decision. The calling agent binds model and reasoning effort from the sam-orchestrate host-runtime-matrix…
Consult Codex as a read-only advisor for a focused assumption, tradeoff, architecture question, security concern, difficult diagnosis, or high-risk decision. The calling agent binds model and reasoning effort from the sam-orchestrate host-runtime-matrix…
Rapidly triage or fully falsify consequential system-development plans through blind specialist reviews, explicit rebuttals, bounded revision rounds, and evidence-weighted decisions. Use for architecture, features, migrations, incidents, releases,…
Run the full task pipeline: sam-plan, sam-refine-task, sam-work delivery, closure review/council, and a proposal-only learning audit. Use when the user runs /sam-task, wants plan-to-PR delivery with final adversarial cleanup, or asks to plan refine implement…
Make a requested interaction feel instantaneous while the real work is still running, using measured immediate feedback, optimistic updates with proven rollback, layout-stable placeholders, streaming, prefetch, and backgrounding — without faking progress,…
Stress-test and revise a technical strategy without implementing it, using an exact evidence baseline, fact and assumption ledgers, bounded loophole analysis, verification mapping, and a validated confidence decision. Use for implementation plans, debugging…
Create and validate risk-based Playwright browser tests for changed or reported user flows, including real linked UI/backend proof, exact route and network assertions, permissions, persistence, and video evidence (mandatory under a parent workflow, local-only…
Design, implement, and validate risk-based regression coverage across unit, component, integration, API/contract, and browser E2E layers, selecting the smallest reliable proof for each changed behavior. Use when asked to add tests, prove a bug fix, increase…
Create or update an evidence-backed description for a pull request, merge request, or equivalent change proposal using the actual base, immutable branch diff, complete file coverage, verified tests and safety claims, and deterministic validation. Use when…
Run an evidence-backed review of local files, staged or unstaged work, branches, commits, diff ranges, pull requests, merge requests, or equivalent remote proposals, with immutable diff coverage, calibrated tests, a validated decision, and optional explicitly…
Record and validate a human-paced MP4 walkthrough of a completed task, feature, or bug fix using the real runnable UI, with acceptance-criterion traceability, privacy checks, deterministic media validation, and optional external publication. Use when asked…
Implement a new codebase capability end-to-end with frozen requirements and scope, risk-calibrated test-first delivery, behavior proof, validated evidence, and optional publication only when explicitly requested. Use for new screens, endpoints, integrations,…
Diagnose and repair broken existing behavior with exact reproduction, proven root cause, frozen scope, the smallest safe correction, calibrated regression proof, and validated evidence. Use for defects, regressions, crashes, incorrect results, failed user…
Simplify code introduced by a completed task while preserving observable behavior, public contracts, unrelated dirty work, and existing proof. Use after implementation already works when asked to reduce duplication, branching, state, indirection, speculative…
Execute a software task through the complete bug-or-feature implementation, refinement, review, simplification, test-coverage, proposal, browser-proof, and demo-video workflow without asking for permission or confirmation on any step. Use when the user wants…