Menu
Review and improve all loaded agent skills based on session patterns and feedback.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Relentless goal/design interview that sharpens framing and writes durable docs (brainstorm artifact, ADRs, glossary) as decisions crystallize. Use when refining a flywheel goal, pressure-testing scope, or "grill with docs".
Start or resume the full agentic coding flywheel. Drives the complete workflow: scan → discover → plan → implement → review.
Set up flywheel prerequisites for this project.
One-shot diagnostic of every flywheel dependency — MCP connectivity, Agent Mail liveness, br/bv/ntm/cm binaries, node version, git status, dist-drift, orphaned worktrees, and checkpoint validity. Use when debugging toolchain issues, before starting a new session, after /flywheel-cleanup, or as a CI gate.
Strategic gap analysis between vision (AGENTS.md / README.md / plan docs) and what's actually implemented. Converts gaps into beads, optionally launches a swarm. Use when "reality check", "where are we really", "gap analysis", "did we drift", or before declaring a long-running project done.
Launch a parallel swarm of agents to implement multiple beads simultaneously.
| name | flywheel-refine-skills |
| description | Review and improve all loaded agent skills based on session patterns and feedback. |
Refine all agent skills.
List all skills in the skills/ directory.
Search agent-mail history for skill-related patterns via search_messages with query "skill feedback" and "planning pattern".
Read current bead completion data from br list --json (closed beads, review feedback).
Setup coordination:
Bootstrap Agent Mail: macro_start_session(human_key: cwd, program: "claude-code", model: your-model, task_description: "Refine all skills").
Create a team: TeamCreate(team_name: "refine-skills").
For each skill found, spawn an analysis agent with run_in_background: true:
Agent(
subagent_type: "general-purpose",
name: "skill-<name>",
team_name: "refine-skills",
run_in_background: true,
prompt: "
Bootstrap Agent Mail: macro_start_session(human_key: '<cwd>', program: 'claude-code', model: 'claude-sonnet-4-6', task_description: 'Refine skill: <name>')
Analyze: given these session patterns and bead outcomes, what improvements would make this skill more effective?
Write proposed changes to docs/skill-refine-<name>-proposed.md.
Send the file path to <your-coordinator-name> via send_message when done.
"
)
Save each task ID for potential TaskStop use. Nudge idle agents individually by name.
After all agents report, shutdown each individually:
SendMessage(to: "skill-<name>", message: {"type": "shutdown_request", "reason": "Analysis complete."}).
Do NOT broadcast to "*".
Present findings per skill with proposed changes (read from the docs files agents wrote).
Ask which skills to update.
For each approved skill, apply changes to the SKILL.md file.
Summarize: "Updated N skills with improvements."