Agent Skills by Emma Hyde
Showing 19 vetted skills indexed across 1 GitHub repositories.
lightpanda-mcp
Scrape web pages and extract data via Lightpanda MCP server.
autoresearch
Automate iterative mutation and validation of pipeline files using YAML and Python.
learn
Stores and retrieves user-specific facts, corrections, and preferences.
hypotheses
Review, promote, reject, or edit LLM-inferred behavioral hypotheses.
health
Evaluate stored memory relevance using scoring, recency, and importance metrics.
breakdown
Analyze memory injection frequency, usage patterns, and importance trends.
reflect
Analyze behavioral data to propose self-model updates for user approval.
evolve
Replay session transcripts through the memesis pipeline to evaluate memory fidelity.
canvas
Load JSONL, markdown, and text files into the review-canvas web UI.
recall
Search stored memories and synthesize conversational summaries from matches.
dashboard
Summarize AI memory system status and performance metrics.
backfill
Import and consolidate past Claude Code conversation transcripts into a memory system.
threads
Visualize interconnected memory clusters as narrative evolution arcs.
memesis reindex --vec
Recompute stored memory embeddings to match a new embedding model.
run-eval
Run retrieval and injection evaluation tests on synthetic and real data.
ideate
Automate scheduled cron triggers and iterative ideation cycles for memesis development.
teach
Decompose complex processes into tagged, linked memory records.
forget
Delete or archive memories by ID or description with audit logging.
connect
Link related memories into cohesive narrative threads using full-text search.