skills
Enhance instruction adherence in CLAUDE.md using <important if> blocks. Learn how explicit relevance signals guide Claude Code, preventing rule deprioritization and improving model performance.
How the Show-Me Skill Decides Between Mermaid Diagrams, ASCII Sketches, and HTML ArtifactsLearn how the show-me skill chooses between Mermaid diagrams, ASCII sketches, and HTML artifacts by selecting the smallest visual representation that conveys the essential point.
How to Use Composite GitHub Actions in build‑iterated‑agentic‑loop and design‑control‑loop for Reusable YAMLDiscover the recommended pattern for using composite GitHub Actions in build-iterated-agentic-loop and design-control-loop for reusable YAML workflows across repositories. Streamline your automation.
How Design-Control-Loop Determines Cadence and Controller Batch Size from Task Risk and Review BurdenLearn how design-control-loop maps task risk and review burden to automation cadence and controller batch size. Ensure safe and reviewable pull requests.
How to Dry-Run a design-control-loop Workflow With No Prior Run HistoryLearn to dry run a design-control-loop workflow without prior history. This guide shows how to use a temporary push trigger and then remove it, preserving your workflow's integrity.
Why Manual workflow_dispatch Runs Bypass the Open-PR Bound Check in Design-Control-Loop Phase GLearn why manual workflow_dispatch runs bypass open-PR bound checks in Design-Control-Loop Phase G. Developers gain override control while automated safety limits remain for scheduled runs.
Open-PR Bound Pattern in Design-Control-Loop Phase G: Workflow Control ImplementationImplement the open PR bound pattern in Phase G for efficient workflow control. Learn how this design enforces single pull requests per agent using GitHub queries and loop labels.
What Belongs in .github/agent-memory/<task-slug>.md Versus the PR ConversationLearn where to store durable agent guidance in .github/agent-memory/<task-slug>.md and transient feedback in PR conversations for a clear two-stage feedback loop.
How the Agent Memory File Is Deterministically Loaded into the Actuator ContextLearn how the agent memory file is deterministically loaded into the actuator context via the --memory flag in GitHub Actions. Discover the process after the controller step.
agent-iteration.ts Reference: Footer and Prompt Modes for /iterate PR CommentsLearn how agent-iteration.ts streamlines PR feedback with footer and prompt modes. It injects metadata and generates execution contexts for iterative coding workflows.
Differences in Secret Names and Response Extraction Paths for Claude Code, Codex, OpenCode, and CodeLayer Agent RunnersCompare secret names and response extraction paths for Claude Code, Codex, OpenCode, and CodeLayer agent runners. Understand API keys and data handling.
How to Template Agent Runners for Headless CI Execution Across Claude Code, Codex, OpenCode, and CodeLayerTemplate agent runners for headless CI execution across Claude Code, Codex, OpenCode, and CodeLayer. Standardize runtime setup, API keys, and output extraction with a single YAML file in humanlayer/skills.
Have a question about this repo?
These articles cover the highlights, but your codebase questions are specific. Give your agent direct access to the source. Share this with your agent to get started:
curl -s "https://instagit.com/install.md" 
Maintain an open-source project? Get it listed too →