Spawns parallel review subagents over session transcripts to route learnings into skill edits.
npx skills add https://github.com/mmdmcy/fluttAIrbar --skill reflect-mmdmcy
Please help me install this Agent Skill. Skill: reflect Source: https://github.com/mmdmcy/fluttAIrbar/tree/main/plugins/pstack/skills/reflect Command: npx skills add https://github.com/mmdmcy/fluttAIrbar --skill reflect-mmdmcy
High-intent search queries and answers about installing and using this skill.
Invoke the reflect skill after a complex task completes. It locates the active transcript, spawns three parallel reviewer subagents with different lenses, synthesizes their findings, and routes approved learnings into edits on existing skills.
Run it after complex tasks with 5+ tool calls that landed cleanly, after the agent hit dead ends and found a generalizable path, or after user corrections. Skip trivial conversations and one-offs that are already covered by existing skills.
No. The synthesizer produces an Accepted/Rejected/Backlog list that is presented to the user for explicit approval before any edit is applied. Substantive edits are handed to the $skill-creator workflow rather than applied directly.
Findings with no existing skill home are routed as new skill proposals via $skill-creator, while items better enforced by lint rules or scripts go to a backlog tracker. Findings that fail durability or specificity criteria are rejected with reasons.
Reviewers need the actual transcript to cite concrete evidence for each learning, including turn numbers and quotes. If no transcript path resolves, a digest of the session is passed to reviewers instead.
itallstartedwithaideaExtract recurring execution patterns and failure modes from agent session transcripts.
firstsun-devCreate, evaluate, and iteratively improve AI agent skills with benchmarked testing.
adjohnConverts recurring corrections into lint rules, runtime checks, and automation scripts.
cpfcoachingExtract reusable problem-solving techniques and patterns from work sessions.
