Command Claude Code
Use when an agent should inspect the shipped Harness Forge command catalog before inventing an execution path.
Command Claude Code
Use when an agent should inspect the shipped Harness Forge command catalog before inventing an execution path.
Command Claude Code
Use when an agent should capture an architecture-significant choice as a durable ASR or ADR inside the Harness Forge runtime.
Command Claude Code
Use when an agent should run a Discover -> Define -> Develop -> Deliver workflow before implementing a meaningful feature.
Command Claude Code
Use when an agent should initialize its Harness Forge working posture, read the compact agent brief, or bootstrap Harness Forge before deeper analysis or implementation.
Command Claude Code
Use when an agent should run repo-intelligence and recommendation flows before setup, architecture, or implementation work.
Command Claude Code
Use when an agent should autonomously escalate a hard task into a recursive session, prefer Typed RLM first, and only fall back to bounded structured analysis when needed.
Command Claude Code
Use when an agent should escalate difficult work into recursive structured-analysis or typed recursive RLM execution with durable session artifacts.
Command Claude Code
Use when an agent should regenerate Harness Forge runtime summaries and repo-intelligence artifacts after install or content changes.
Command Claude Code
Use when an agent should review runtime health, drift, decision coverage, and stale task artifacts in an installed Harness Forge workspace.
Command Claude Code
Use when an agent should inspect the current Harness Forge workspace state before acting.
Command Claude Code
Use when an agent should inspect Harness Forge task-runtime folders, task packs, and linked artifacts for the current work item.
Command Claude Code
Use when an agent should preview or apply a non-destructive Harness Forge package refresh while preserving gathered runtime state.
Command Claude Code
Use when a task needs structure, sequencing, risks, and validation checkpoints.
Command Claude Code
Use when defining or running verification for code, docs, templates, or migration work.
Skill Claude CodeCodex
Review and stage dependency upgrades with compatibility and rollback awareness.
Skill Claude CodeCodex
Skill "engineering-assistant" from ldilov/harness-forge, covering engineering assistant, trigger signals, inspect first, workflow and output contract.
Skill Claude CodeCodex
Rapid incident framing for breakages, production regressions, and severe failures.
Skill Claude CodeCodex
Use this skill when local events, derived summaries, or benchmark expectations need to be inspected or updated.
Skill Claude CodeCodex
Define local-first logging, metrics, and diagnostic signals for a runtime surface.
Skill Claude CodeCodex
Profile likely hotspots and turn performance work into measurable next steps.
Skill Claude CodeCodex
Turn a change set into a concise implementation and risk summary.
Skill Claude CodeCodex
Use this skill when you need a repo-aware map before recommending new guidance surfaces.
Skill Claude CodeCodex
Define right-sized test layers and coverage priorities for a change.
Agent Claude Code
Starter prompt for documentation lookup.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: