Open Horizon Labs
Official@open-horizon-labs · United States of America
The other half of AI
Agent Skills by Open Horizon Labs
Showing 24 vetted skills indexed across 3 GitHub repositories.
superego
Evaluate coding conversations for alignment, focus, and proportionality before edits.
problem-statement
Define and reframe problems into structured, testable problem statements.
dissent
Surface contrary evidence and assumptions during major decisions.
salvage
Extract structured learnings and restart context from drifting sessions.
solution-space
Generate and compare candidate solution approaches with trade-offs and a local-maxima check.
review
Detect drift between intent and execution at natural pause points.
teach-oh
Generate AGENTS.md files with structured Open Horizons Framework context across sessions.
execute
Execute chosen solutions with pre-flight checks, incremental builds, drift detection, and salvage handling.
ship
Coordinate merged code delivery through CI/CD pipelines and production readiness checks.
aim
Define a concrete user behavior change with current and desired states.
problem-space
Identify project objectives, constraints, and terrain into a Problem Space Map.
bottle
Manage Open Horizon Labs tools via a scriptable CLI.
wm
Manage AI session memory with wm CLI commands like dive, compile, and distill.
sg
Evaluate AI decisions with sg CLI commands like review, init, and audit.
ba
Track AI-session tasks with file-based CLI commands in .ba/ directories.
dive-prep
Gather local documents and repository state into a .wm/dive_context.md file.
oh-plan
Convert grounding context into 1–4 GitHub issues with acceptance criteria.
drummer
Batch review and merge stacked PRs labeled drummer-merge with CI checks and rebasing.
oh-notes
Manage GitHub PR feedback by creating isolated worktrees and guiding fixes.
mouse
Automate ba task workflows with Git branches and PR orchestration.
jira-plan
Convert session context into Jira subtasks under a specified parent.
notes
Orchestrate a focused worktree workflow to address pull request feedback.
oh-task
Automate GitHub issue completion from branch worktree to pull request.
oh-merge
Batch-review and merge stacked GitHub PRs in dependency order.
Frequently Asked Questions About Open Horizon Labs
FAQPage SchemaWhat specific tasks can engineers perform using these capabilities?▼
Engineers can manage complex pull request feedback loops, batch-merge stacked branches, track session-based tasks via local file directories, and convert project context into structured GitHub issues or Jira subtasks.
Which personas benefit most from these technical methodologies?▼
Senior software engineers, technical leads, and engineering managers who oversee complex, multi-branch development cycles and require rigorous decision-tracking to prevent drift between project intent and final code execution.
What are the prerequisites for implementing these methodologies?▼
Implementation requires a Git-based environment, familiarity with worktree-based development, and existing integration with GitHub or Jira for issue tracking. Users must be comfortable managing local configuration directories to maintain session state.