Mike Stop Continues
Community@mikestopcontinues
Fan of freedom, creativity, and code. Serial founder. Currently @facebook. Formerly @mongodb. Exited @browsercat.
Agent Skills by Mike Stop Continues
Showing 21 vetted skills indexed across 1 GitHub repositories.
mikestopcontinuesyolo-project
Classifies ambiguous prompts and routes them to the matching lifecycle skill.
mikestopcontinuescreate-note
Create living notes for unresolved decisions and design context via a scaffold CLI.
mikestopcontinuescreate-spike
Orchestrates a full research spike lifecycle by composing document authoring, validation, triage, and decision-locking skills.
mikestopcontinueslock-decisions
Walks decision-shaped findings through lock, defer, or grill prompts and records outcomes.
mikestopcontinuescreate-decision
Writes captain-locked decisions as numbered dNN markdown files in project or global scopes.
mikestopcontinuesreview-code
Reviews a code diff through one focus brief and writes a structured severity-tiered review file.
mikestopcontinuesexecute-plan
Executes approved plans phase-by-phase using worktrees, code validation, and feedback triage cycles.
mikestopcontinuestriage-feedback
Clusters, dedupes, and routes multi-reviewer feedback findings for an artifact.
mikestopcontinuescreate-plan
Orchestrates plan creation with per-chapter authoring and iterative validation loops.
mikestopcontinueslaunch-project
Loads or scaffolds project context from cwd or prompt for lifecycle skills.
mikestopcontinuesreview-doc
Reviews one markdown artifact through a single focus and writes a structured rNN-{focus}.md findings file.
mikestopcontinueswrite-convention
Creates, edits, or deletes convention files under docs/conventions/ and syncs the AGENTS.md convention list.
mikestopcontinuesarchive-project
Condenses a completed project directory into a single archived README with promoted decisions and rewritten cross-references.
mikestopcontinuesvalidate-doc
Dispatches parallel review-doc subagents to validate markdown documents across multiple focus areas.
mikestopcontinuesrevise-doc
Applies review findings from rNN- review files to a document during validate-loop iteration.
mikestopcontinuesprocess-feedback
Resolves one feedback cluster against primary sources and returns a canonical verdict.
mikestopcontinuesvalidate-code
Dispatches parallel review-code subagents against a code diff and summarizes results.
mikestopcontinueswrite-code
Implements code changes in isolated git worktrees with incremental commits and quality-gate enforcement.
mikestopcontinuesstart-session
Detects worktree state and project context to summarize recent work and suggest next steps.
mikestopcontinuesgrill-me
Conducts a Socratic interview to resolve open design decisions one question at a time.
mikestopcontinueswrite-doc
Author and edit Markdown documents under docs/ following project conventions and scaffold rules.
Frequently Asked Questions About Mike Stop Continues
FAQPage SchemaWhat tasks can I accomplish with Mike Stop Continues' skills?▼
You can scaffold and load projects, author plans and spikes, write and review code in git worktrees, run parallel document and code reviews, triage feedback clusters, lock decisions as numbered records, maintain conventions, and archive finished projects with atomic commits.
Who are these skills designed for?▼
They target developers ('captains') running agent-assisted coding sessions who want structured lifecycle discipline: intent routing, per-phase plan execution, persona-based review subagents, and persistent decision capture rather than ad-hoc prompting.
How do the skills operate at runtime?▼
Skills invoke each other via /skill-name chaining. yolo-project routes intent, launch-project loads context, and lifecycle skills like create-plan or execute-plan orchestrate validate, triage, revise, and lock-decisions loops using Read, Write, Bash, and Dispatch primitives.
What prerequisites do these skills require?▼
A git-based project environment with worktree support, a docs/ directory structure for plans, decisions, notes, and conventions, and an agent runtime supporting the declared primitives (Read, Edit, Write, Glob, Grep, Bash, WebFetch, WebSearch, TodoWrite, Dispatch).
How are decisions and reviews persisted?▼
Locked decisions are saved as numbered dNN-<name>.md files via create-decision, while every review produces a single-focus rNN-{focus}.md file beside the artifact, enabling triage-feedback to dedupe, cluster, and route findings to revision or locking.