Results for “fork-mode”

Provides guidance for enterprise-grade RL training using miles, a production-ready fork of slime. Use when training large MoE models with FP8/INT4, needing train-inference alignment, or requiring speculative RL for maximum throughput.

OI-3 spike harness — the FORKED (context fork) readability-guidance variant. Same payload as the inline variant but declared to run in a forked context. Use only when a spike consumer skill invokes it.

Split an idea or active research path into two or more new research product-path stubs after alignment approval; preserves the source path by default and archives it only when --archive is passed

Version 2.35.0 | PRD to Production | Zero Human Intervention > Research-enhanced: OpenAI SDK, DeepMind, Anthropic, AWS Bedrock, Agent SDK, HN Production (2025)

Automates the entire software development lifecycle from PRD to production with zero human intervention, using multi-agent orchestration and memory systems.

Version 2.35.0 | PRD to Production | Zero Human Intervention > Research-enhanced: OpenAI SDK, DeepMind, Anthropic, AWS Bedrock, Agent SDK, HN Production (2025)

Full safety mode: destructive command warnings + directory-scoped edits. (gstack)

Full safety mode: destructive command warnings + directory-scoped edits. (gstack)

Fork, sanitize, and package private projects for safe public release through a three-stage pipeline.

Runs an autonomous multi-agent software development pipeline that takes a PRD through to production with zero human intervention, using model-tiered agents, memory, and verification cycles.

Version 2.35.0 | PRD to Production | Zero Human Intervention > Research-enhanced: OpenAI SDK, DeepMind, Anthropic, AWS Bedrock, Agent SDK, HN Production (2025)

Full safety mode: destructive command warnings + directory-scoped edits. (gstack)

OI-3 spike harness — heavy consumer skill, FORKED arm. Builds an incident post-mortem and sources the readability standard mid-workflow via a forked (context fork) guidance skill. Use only when explicitly invoked by the spike harness with a TRIAL_ID and data path.

Runs focused experiments to validate feasibility of an idea, saving artifacts to .planning/spikes/ and supporting both idea and frontier modes.

Ultra-compressed communication mode. Cuts token usage ~75% by dropping filler, articles, and pleasantries while keeping full technical accuracy. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman.

Multi-agent autonomous startup system for Claude Code. Triggers on "Loki Mode". Orchestrates 100+ specialized agents across engineering, QA, DevOps, security, data/ML, business operations, marketing, HR, and customer success. Takes PRD to fully deployed, revenue-generating product with zero human intervention. Features Task tool for subagent dispatch, parallel code review with 3 specialized reviewers, severity-based issue triage, distributed task queue with dead letter handling, automatic deployment to cloud providers, A/B testing, customer feedback loops, incident response, circuit breakers, and self-healing. Handles rate limits via distributed state checkpoints and auto-resume with exponential backoff. Requires --dangerously-skip-permissions flag.

Use when the user explicitly enters TURBO, mode TURBO, enable TURBO, or TURBO for this task, and Codex should optimize for maximum quality without normal token-cost restraint.

Provides a frontend design mode for strong visual direction, UI polish, redesigns, landing pages, and production-grade interface work, including shadcn-based surfaces.

Compresses AI agent output tokens by up to 65% using a terse, article-free communication style while preserving full technical accuracy. Supports multiple intensity levels and auto-disables for security warnings or ambiguous instructions.

Design task-local harnesses, eval gates, and reusable skill extraction for adaptive agent workflows.

Narrates a demo video for Loki Mode, a multi-agent autonomous startup system for Claude Code, covering setup, phases, and features.

Maintains a working memory file for AI agents, tracking active goals, completed tasks, performance metrics, blockers, and key decisions across sessions.

CEO/founder-mode plan review. Rethink the problem, find the 10-star product, challenge premises, expand scope when it creates a better product. Four modes: SCOPE EXPANSION (dream big), SELECTIVE EXPANSION (hold scope + cherry-pick expansions), HOLD SCOPE (maximum rigor), SCOPE REDUCTION (strip to essentials). Use when asked to "think bigger", "expand scope", "strategy review", "rethink this", or "is this ambitious enough". Proactively suggest when the user is questioning scope or ambition of a plan, or when the plan feels like it could be thinking bigger. (gstack)

Codify the most recent successful /scrape flow into a permanent browser-skill on disk. Future /scrape calls with the same intent run the codified script in ~200ms instead of re-driving the page. Walks back through the conversation, synthesizes script.ts + script.test.ts + fixture, runs the test in a temp dir, and asks before committing. Use when asked to "skillify", "codify", "save this scrape", or "make this permanent". (gstack)

Execute remote commands on Windows targets using WMI-based lateral movement techniques, including Impacket wmiexec.py, CrackMapExec, and native PowerShell WMI commands for red team engagements.

Fast headless browser for QA testing and site dogfooding. Navigate pages, interact with elements, verify state, diff before/after, take annotated screenshots, test responsive layouts, forms, uploads, dialogs, and capture bug evidence. Use when asked to open or test a site, verify a deployment, dogfood a user flow, or file a bug with screenshots. (gstack)

Runs and writes tests, verifies behavior, and orchestrates two-pass TDD across scoped modes, blocking on quality gates and never editing production code.

Use the status-briefing skill to run `pk-reconcile session-start`, then

Capture friction as methodology notes. Three modes — explicit description, contextual (review recent corrections), session mining (scan transcripts for patterns). Triggers on "/remember", "/remember [description]".

Full safety mode: destructive command warnings + directory-scoped edits. Combines /careful (warns before rm -rf, DROP TABLE, force-push, etc.) with /freeze (blocks edits outside a specified directory). Use for maximum safety when touching prod or debugging live systems. Use when asked to "guard mode", "full safety", "lock it down", or "maximum safety". (gstack)

Multi-agent autonomous startup system for Claude Code. Triggers on "Loki Mode". Orchestrates 100+ specialized agents across engineering, QA, DevOps, security, data/ML, business operations, marketing, HR, and customer success. Takes PRD to fully deployed, revenue-generating product with zero human intervention. Features Task tool for subagent dispatch, parallel code review with 3 specialized reviewers, severity-based issue triage, distributed task queue with dead letter handling, automatic deployment to cloud providers, A/B testing, customer feedback loops, incident response, circuit breakers, and self-healing. Handles rate limits via distributed state checkpoints and auto-resume with exponential backoff. Requires --dangerously-skip-permissions flag.

Implements code features phase by phase from a JSON or Markdown plan, with configurable unit-test and quality-gate choices, token-budgeted execution, and resumable state.

Land and deploy workflow. (gstack)

Multi-agent autonomous startup system for Claude Code. Triggers on "Loki Mode". Orchestrates 100+ specialized agents across engineering, QA, DevOps, security, data/ML, business operations,...