burningportra
Community@burningportra
burningportra maintains a Claude Code skill registry for agentic coding flywheels, swarm orchestration, session memory, debugging, and code review.
Agent Skills by burningportra
Showing 41 vetted skills indexed across 1 GitHub repositories.
burningportraflywheel-refine-skills
Reviews and improves agent skills using session patterns and bead completion feedback.
burningportraflywheel-swarm-stop
Stops all running swarm agents and releases their file reservations.
burningportraflywheel-cleanup
Detects and removes orphaned git worktrees left by crashed or stopped sessions.
burningportraflywheel-stop
Stops the active flywheel session and clears checkpoint state.
burningportragdb-for-debugging
Diagnose process hangs, crashes, and deadlocks using GDB, strace, and /proc inspection.
burningportramemory
Search, store, and prune CASS long-term memory entries via CLI and MCP operations.
burningportra_template
Provides a boilerplate SKILL.md and slash-command template for authoring new agent-flywheel skills.
burningportraflywheel-scan
Scans targeted repository paths for architecture, performance, security, testing, or dependency concerns.
burningportraflywheel-swarm
Launch parallel NTM-managed agents to implement multiple beads simultaneously.
burningportracass
Search and mine past agent session logs for prompts, decisions, and patterns.
burningportraflywheel-status
Displays flywheel checkpoint state, bead progress, inbox messages, and next recommended bead.
burningportraslb
Enforces two-person peer review for destructive shell commands from AI agents.
burningportracass-memory
Builds persistent cross-agent procedural memory from coding session history.
burningportraflywheel-reality-check
Analyzes gaps between project vision documents and actual implementation, converting findings into tracked beads.
burningportraflywheel-fix
Applies targeted code fixes through a single-bead agent workflow with review.
burningportraflywheel-doctor
Diagnose flywheel toolchain dependencies and render a per-check health report with inline remediation.
burningportrastart
Orchestrates the full agentic coding flywheel from goal intake through implementation and review.
burningportraflywheel-setup
Installs and configures agent-flywheel prerequisites including br, bv, ntm, and agent-mail.
burningportraflywheel-healthcheck
Audits flywheel dependencies, codebase state, and bead graph into a scored health report.
burningportraflywheel-rollback
Reverts a completed bead implementation using git revert and restores its open status.
burningportradsr
Builds and releases cross-platform binaries locally when GitHub Actions is throttled.
burningportraui-ux-polish
Generates numbered UI/UX improvement suggestions and converts selected items into implementation beads.
burningportraflywheel-tool-feedback
Submit structured tool feedback to agent-mail for improving flywheel sessions.
burningportraflywheel-duel
Runs state-aware multi-agent adversarial duels and routes artifacts into flywheel pipelines.
Frequently Asked Questions About burningportra
FAQPage SchemaWhat tasks can I accomplish with burningportra's skill collection?▼
You can run full agentic coding flywheels (scan, discover, plan, implement, review), launch parallel agent swarms, audit codebases for bugs and security issues, debug stuck processes with GDB and strace, mine session history for reusable patterns, polish UI/UX, rebuild changelogs, and cross-validate decisions across multiple models.
Who is the target audience for these skills?▼
The skills target engineers running Claude Code agent sessions on real codebases: developers coordinating multi-agent swarms, maintainers needing gap analysis between plans and implementation, and operators handling destructive commands, releases, and process-level debugging in terminal-driven environments.
How do the flywheel skills work in practice?▼
Start with flywheel-setup to configure prerequisites, then run the start skill to drive the scan-discover-plan-implement-review cycle. Use flywheel-doctor to verify MCP connectivity, binaries, and git state; flywheel-status tracks bead progress; flywheel-stop and flywheel-cleanup reset sessions and orphaned worktrees.
What prerequisites and dependencies do these skills require?▼
Skills run inside Claude Code as a plugin namespace (agent-flywheel). The flywheel-doctor skill checks required dependencies: MCP connectivity, Agent Mail liveness, br/bv/ntm/cm binaries, node version, git status, and checkpoint validity. Destructive operations use the slb two-person rule requiring peer review.
Are these skills open source and what licensing applies?▼
The frontend-design skill is explicitly Apache 2.0 licensed, based on Anthropic's frontend-design skill with attribution in NOTICE.md. Other skills in the manifest do not declare licenses; the collection is published as public repositories under the burningportra GitHub account.