Analyze completed plan diffs and route learnings to agent templates and docs.
npx skills add https://github.com/danalexilewis/taskgraph --skill evolve-danalexilewis
Please help me install this Agent Skill. Skill: evolve Source: https://github.com/danalexilewis/taskgraph/tree/main/.cursor/skills/evolve Command: npx skills add https://github.com/danalexilewis/taskgraph --skill evolve-danalexilewis
Post-plan pattern mining to review completed plan execution for anti-patterns and route learnings to agent templates and docs. It enables a targeted post-mortem that informs templates like implementer.md and quality-reviewer.md and updates to docs/skills.
Invoke evolve after a plan completes to extract diffs and route learnings to templates and docs.
High-intent search queries and answers about installing and using this skill.
Analyzing plan diffs for anti-patterns involves inspecting completed plan execution changes to surface implementation issues. It extracts structural diffs from the most recent plan and generates structured findings to inform cross-team learning and post-mortem reviews.
Plan-diff analysis is a post-plan pattern mining technique that reviews completed plan changes for implementation anti-patterns. You need it after a plan completes to route extracted learnings to agent templates and docs, enabling targeted post-mortems for cross-team development.
Routing learnings to agent templates and docs is achieved by analyzing plan diffs and generating structured findings. The process targets templates like implementer.md and quality-reviewer.md, updating durable knowledge across docs/skills for future use.
Plan-evolve workflows support multi-agent orchestration prior to the plan-merge step. By invoking diff analysis after a plan completes, the process returns structured findings that route learnings to agent templates and docs for action.
Analyzing plan diffs for anti-patterns targets only the most recent plan execution and requires a completed plan to extract changes. It focuses strictly on post-plan pattern mining and routing learnings to templates, not pre-execution validation.
anotbGenerate a structured regulatory remediation plan with workstreams, milestones, and governance checkpoints.
mupengi-botCollect subagent learning reports and integrate insights into memory and knowledge bases.
JubaKitiashviliCoordinate multi-agent pipelines across planning, implementation, testing, review, and validation phases.
dyyounggCapture CUDA crash boundaries and tensor metadata via kernel API logging in SGLang.
