chaotic_semantic_memory

GOAP-based orchestrator for managing GitHub issues, creating action plans, and executing workspace operations with branch/PR workflow.

Configure and troubleshoot npm OIDC Trusted Publishing for GitHub Actions. Use when npm publish fails with E404, OIDC errors, or provenance issues.

GitHub release management, crates.io trusted publishing, npm provenance, and GitHub Pages documentation. Use when creating releases, publishing packages, or deploying docs.

Git commit conventions, validation gates, and CI/CD workflows. Use when committing changes, verifying gates, or working with GitHub Actions.

Validate merge readiness with atomic commits and GitHub Actions checks using gh CLI; use for pre-merge verification and CI truth validation.

Apply TRIZ inventive principles to solve problems that seem to have no good solution. Use after triz-analysis has identified contradictions.

Implement or refactor Rust in this repository. Use when writing new modules, modifying existing source files, or adding features to the chaotic_semantic_memory crate.

Orchestrate static analysis using Codacy CLIs. Use when Codacy blocks a PR, for fixing quality gate failures, or suppressing false positives in this Rust/Shell codebase.

Portable verification skill for memory lifecycle operations (save, load, archive, delete) with file-level and database-level checks.

Run and analyze criterion benchmarks for performance-sensitive changes. Use when optimizing hot paths, validating perf targets, or comparing baselines.

Test-fix-validate loops for complex changes: run tests, identify failures, fix, validate, repeat until green. Use for red-green-refactor cycles.

Automated CI fix cycle: detect failure, classify error, apply fix, retry. Use after push triggers CI failure for automated repair iteration.

Create new skills with proper structure and evals. Use when creating domain-specific skills for the .agents/skills/ directory.

Internal dogfooding skill for storing and recalling engineering context through the csm CLI during day-to-day development.

SIMD optimization, connection pooling, batch APIs, and caching. Use when improving throughput or reducing latency.

Property-based testing, fuzzing, and edge case coverage. Use when adding comprehensive test coverage with proptest or cargo-fuzz.

Validate the chaotic_semantic_memory crate: compile, test, lint, LOC caps, and benchmarks. Use when asked to validate, check, or verify the build.

Orchestrate GitHub issues and automated tasks via the Jules CLI remote interface.

Capture non-obvious session learnings to project memory. Use at end of session or after solving hard problems to persist insights for future sessions.

Ensure shell scripts pass ShellCheck linting and have BATS tests for complex logic. Use when writing or modifying shell scripts in scripts/ directory.

Evaluate skill performance with benchmarks and metrics. Use when validating skill effectiveness or measuring improvement from skill usage.

Break complex tasks into atomic, dependency-ordered goals. Use for multi-step tasks with unclear boundaries or when planning agent team assignments.

Perform systematic contradiction analysis before implementation. Use when facing design tradeoffs, architectural decisions, or problems where improving one parameter worsens another.

Artifact-aware decision logic for selecting the correct distribution channel (cargo crate vs WASM npm vs CLI npm). Use when validating, installing, or publishing.

Verify memory persistence with Turso/libSQL databases before releases. Use before any release to ensure import/export functionality works correctly, data roundtrips successfully, and no serialization regressions.

Create high-level architecture diagrams using draw.io. Use when asked to create, visualize, or diagram system architecture, data flows, or module relationships.

Multi-persona code analysis orchestrator using RYAN, FLASH, and SOCRATES for balanced decision-making. Invoke for complex architectural decisions or trade-off analysis.

Build ordered, executable action plans from current state to target state using explicit preconditions, effects, and costs.

Tracing, metrics, derive macros, and error context. Use when adding observability or improving developer experience.

Export/import, versioning, migrations, and backup/restore. Use when adding enterprise/production features.

Debug and tune the echo state network reservoir. Use when diagnosing spectral radius issues, chaotic dynamics problems, sparse weight anomalies, or reservoir-to-hypervector projection failures.

Write or update ADRs for architecture-impacting changes, major tradeoffs, or decisions requiring durable rationale and consequences.