rust-self-learning-memory

Checkpoint episodes and create handoff packs for multi-agent session continuity

Configure, test, benchmark, and manage embedding providers for semantic search and vector operations

Create and traverse typed relationships between episodes for dependency tracking and knowledge graphs

Manage episode tags for organization, filtering, and retrieval across the memory system

Generate task playbooks and explain patterns using learned strategies from episodic memory

Record and analyze recommendation sessions and feedback to drive the self-learning loop

Use and troubleshoot the Memory MCP server for episodic memory retrieval and pattern analysis. Use when working with MCP server tools, validating the MCP implementation, or debugging MCP server issues.

Integrate external signal providers (AgentFS, audit trails, toolcall logs) into the reward system. Use when adding external reward signals, processing toolcall audit trails, or connecting third-party agent telemetry to episode scoring.

Create new Claude Code skills with proper structure, YAML frontmatter, and best practices. Use when creating reusable knowledge modules, adding specialized guidance, or building domain-specific expertise.

Coordinate multiple specialized Skills and Task Agents through parallel, sequential, swarm, hybrid, or iterative execution strategies. Use when orchestrating multi-worker workflows, managing dependencies, or optimizing complex task execution with quality gates.

Update AGENTS.md and agent_docs/ following best practices. Use when modifying agent guidelines, adding new documentation, or restructuring agent instructions.

Multi-perspective code analysis using three AI personas (RYAN, FLASH, SOCRATES) for comprehensive decision-making. Use when complex code decisions need analysis from multiple viewpoints, or when avoiding single-perspective blind spots is critical.

Dynamically validate that the implemented codebase matches architectural decisions documented in plan files. Use when validating implementation matches planning documents, checking for architecture drift, or preparing for architecture reviews.

Build Rust code with proper error handling, optimization, and workspace support for development, testing, and production

Poll GitHub CI status with exponential backoff until all checks complete. Use after pushing to a PR branch to monitor CI until merge-ready or failure. Prefer gh pr checks --watch and gh run watch over bare sleep loops.

Analyze implementation details, trace data flow, explain technical workings, locate files, and consolidate codebases. Use when you need to understand HOW code works, find file locations, or assess technical debt.

Systematic debugging approach for Rust async code with Tokio, Turso, and redb. Use when diagnosing runtime issues, performance problems, async deadlocks, database connection issues, or panics.

Execute and troubleshoot do-memory-cli commands for episode management, pattern analysis, and storage operations. Use this skill when running CLI commands, debugging CLI issues, explaining command usage, or guiding users through CLI workflows.

Systematic approach to implementing new features in the Rust memory system following project conventions. Use when adding new functionality with proper testing and documentation, maintaining code quality and test coverage.

Manage git worktrees for efficient multi-branch development. Use when creating worktrees for feature branches, organizing worktree directories, cleaning up unused worktrees, or implementing worktree-based workflows.

Background reference for release prep (SemVer, changelog categories). For actual releases, ALWAYS use the release-guard skill and release-manager.sh ship — never manual gh release create.

Diagnose, fix, and optimize GitHub Actions workflows for Rust projects. Use when setting up CI/CD, troubleshooting workflow failures, optimizing build times, or ensuring best practices.

Extract non-obvious session learnings into scoped AGENTS.md files

Execute workflow agents iteratively for refinement and progressive improvement until quality criteria are met. Use when tasks require repetitive refinement, multi-iteration improvements, progressive optimization, or feedback loops until convergence.

Retrieve relevant context from memory and preserve essential state. Use for episode retrieval, semantic search, or context compaction when window fills.

Universal agent memory harness — record, replay, and benchmark real agent sessions. Use when testing memory system learning, generating test fixtures, or benchmarking CLI performance.

Benchmarking and performance optimization for Rust. Use when profiling CPU/memory bottlenecks, running Criterion benchmarks, validating performance improvements, or detecting regressions in CI.

Analyze gaps between implementation plans and actual codebase implementation for the Rust self-learning memory project

Monitor release cadence, detect drift, and coordinate resolution using GOAP orchestrator with swarm agents. Use when release drift is detected, PRs need release-preparation labels, or automated release coordination is required.

Synchronize memories between Turso (durable) and redb (cache) storage layers. Use when cache appears stale, after failures, or during periodic maintenance.

Systematic approach to diagnosing and fixing failing tests in Rust projects. Use when tests fail and you need to diagnose root causes, fix async/await issues, handle race conditions, or resolve database connection problems.

Unified testing patterns for Rust: unit testing quality, episodic memory operations, and async/tokio code. Use when writing tests, reviewing test code, or diagnosing test failures.

Comprehensive PR health check: merge state, CI status, conflicts, cancelled checks, AND all PR comments/reviews (human + bots). Requires addressing actionable feedback before recommending merge. Prevents 'CI green, ready to merge' when conflicts, pending comments, or Codecov/Codacy findings remain.

Resolve queries or URLs into compact, LLM-ready markdown using a low-cost cascade. Prioritizes llms.txt for structured docs, uses web fetch/search tools for extraction. Use when you need to fetch documentation, resolve web URLs to markdown, search for technical content, or build context from web sources.

Diagnose and fix GitHub Actions CI failures for Rust projects. Use when CI fails, tests timeout, or linting issues occur. Captures common patterns from CLAUDE_INSIGHTS_REPORT.md.

Maintain high code quality through formatting, linting, static analysis, and clean code principles. Use for rustfmt, clippy, cargo audit, code reviews, refactoring, and quality gates.

Git commit with enforced quality gates, proper message format, and safe push workflow

Invoke for complex multi-step tasks requiring intelligent planning and multi-agent coordination. Use when tasks need decomposition, dependency mapping, parallel/sequential/swarm/iterative execution strategies, or coordination of multiple specialized agents.

Execute Rust tests (unit, integration, doc). Use cargo nextest for fast parallel execution.

Canonical release workflow for this repo. One path every time: main green → release-manager ship → tag vX.Y.Z → release.yml. Use when the user says release, tag, publish, deploy, version, or cut a GitHub release.