raven

Plugin raven from marketplace raven in giggsoinc/raven. Commands read from the repo's manifest — 24 days ago.

Skills (63)

• router

  Toggle the Raven model router. /router enables advisory Haiku-subagent delegation for simple prompts, /router off disables it, /router status shows the current mode. The primary session model is never changed — that is set only via /model.

• agent-chaining

  Use when designing or reviewing multi-agent systems, subagent orchestration, parallel vs sequential agent execution, context passing between agents, tool delegation, and agent memory patterns. Platform-neutral — covers production multi-agent architecture across AI runtimes.

• aiml-specialist

  Use for any AI/ML Engineering question. Assumes Andrej Karpathy (AI researcher) persona. Deep multi-dimensional analysis. Bullets not prose.

• andie

  USE PROACTIVELY whenever the user asks for: planning, design, architecture decision, tradeoff analysis, comparing approaches, strategy, system design, refactor scope, deciding what to build, or any non-trivial request needing clarification. Also USE when user says 'should I', 'how do I approach', 'p

• aws-specialist

  Use for any AWS question. Assumes Werner Vogels (CTO, Amazon) persona. Deep multi-dimensional analysis. Bullets not prose.

• azure-specialist

  Use for any Azure question. Assumes Mark Russinovich (CTO, Microsoft Azure) persona. Deep multi-dimensional analysis. Bullets not prose.

• bigdata-specialist

  Use for any Big Data / Analytics question. Assumes Matei Zaharia (Spark creator) persona. Deep multi-dimensional analysis. Bullets not prose.

• dataeng-specialist

  Use for any Data Engineering question. Assumes Joe Hellerstein (database researcher) persona. Deep multi-dimensional analysis. Bullets not prose.

• db-specialist

  Universal database orchestrator. Use for ANY database question — schema

• devops-specialist

  Use for any DevOps / SRE question. Assumes Kelsey Hightower (SRE, Google) persona. Deep multi-dimensional analysis. Bullets not prose.

• dynamic-specialist

  Use for ANY platform, library, or domain with no curated specialist

• fastapi-specialist

  Use for any FastAPI question. Assumes Sebastián Ramírez (FastAPI creator) persona. Deep multi-dimensional analysis. Bullets not prose.

• gcp-specialist

  Use for any GCP question. Assumes Jeff Dean (Google Brain) persona. Deep multi-dimensional analysis. Bullets not prose.

• k8s-specialist

  Use for any Kubernetes question. Assumes Kelsey Hightower (K8s evangelist, Google) persona. Deep multi-dimensional analysis. Bullets not prose.

• kafka-specialist

  Use for any Kafka / Streaming question. Assumes Jay Kreps (Kafka creator, Confluent) persona. Deep multi-dimensional analysis. Bullets not prose.

• log-management-specialist

  Use when designing or reviewing logging pipelines, structured logging, log aggregation (ELK, Loki, CloudWatch), retention policies, PII scrubbing, alerting on log patterns, or distributed tracing integration. Expert-level guidance on observability discipline.

• nicegui-specialist

  Use for any NiceGUI question. Assumes Falko Schindler (NiceGUI creator) persona. Deep multi-dimensional analysis. Bullets not prose.

• oci-specialist

  Use for any OCI (Oracle Cloud) question. Assumes Larry Ellison (Oracle founder) persona. Deep multi-dimensional analysis. Bullets not prose.

• odoo-specialist

  Use for any Odoo question — module development, ORM, XML views, QWeb,

• postgres-specialist

  Use for any PostgreSQL question. Assumes Michael Stonebraker (database architecture) persona. Deep multi-dimensional analysis. Bullets not prose.

• raven-core

  Use when writing code, adding imports, creating files, committing,

• raven-document

  Use when adding or updating documentation. Enforces docstrings

• raven-expert

  Use when deep expertise is needed — complex architecture decisions,

• raven-plan

  Use when planning any new feature, module, service, or refactor.

• raven-refactor

  Use when refactoring code to meet Raven standards. Combines

• raven-review

  Use when reviewing code, PRs, or any file changes. Extends

• raven-security

  Use when doing security review of code, PRs, or before any merge.

• raven-test

  Use when writing or reviewing tests. Forces test-first discipline —

• redis-specialist

  Use for any Redis / Caching question. Assumes Antirez (Redis creator, Salvatore Sanfilippo) persona. Deep multi-dimensional analysis. Bullets not prose.

• salesforce-specialist

  Use for any Salesforce question — Apex, SOQL, LWC, Flows, integrations,

• security-specialist

  Use for any Application Security question. Assumes Bruce Schneier (security technologist) persona. Deep multi-dimensional analysis. Bullets not prose.

• task-observer

  Meta-skill. Loaded at every session start via CLAUDE.md boot sequence.

• terraform-specialist

  Use for any Terraform / IaC question. Assumes Mitchell Hashimoto (Terraform creator) persona. Deep multi-dimensional analysis. Bullets not prose.

• tools-landscape

  Tool and product registry for Andie pre-flight. Maps problem types to tool options with tradeoffs. Covers LLMs, vector DBs, streaming, orchestration, IaC, monitoring, secrets, CI/CD, search, auth, open source infra, and cloud platforms. Load when a tool selection decision is in scope.

• vault-specialist

  Use for any HashiCorp Vault / Secrets question. Assumes Armon Dadgar (Vault creator) persona. Deep multi-dimensional analysis. Bullets not prose.

• vector-db-specialist

  Use for any Vector Databases / Search question. Assumes Dmitry Baranchuk (HNSW researcher) persona. Deep multi-dimensional analysis. Bullets not prose.

• andie-frames

  3-phase browser testing skill. Generates UI·UX·Data·Security·Correctness·Speed test cases, drives visible Chrome live, writes Raven-compatible blocker MDs. Human watches and guides throughout. Andie packed inside — no external skill calls.

• andie-guru

  USE PROACTIVELY whenever the user asks: 'explain', 'what does this mean', 'in plain English', 'break it down', 'simplify', 'translate for non-tech', 'help me understand', or asks for a layman summary of any prior Raven/specialist output. Feynman-style explainer. Returns 50-word Feynman + 100-word su

• andie-jr

  USE PROACTIVELY whenever the user reports: a bug, error, exception, stack trace, regression, test failure, 'not working', 'broken', 'why is X failing', 'X stopped working', 'doesn't behave', unexpected output. Brownfield debug assistant. Max 2 rounds, max 3 roles, returns problem · root cause · fix

• db-router

  Pure routing table. Detects DB from context, hands off to the right curated specialist or dynamic-specialist. Carries zero specialist content itself.

• graph-db-specialist

  Use for graph database design, querying, and deployment. Sub-modes — falkordb · neo4j · property-graph. Assumes Jim Webber (Neo4j co-author) persona. Bullets not prose.

• ml-ops-specialist

  Use for ML pipelines, feature stores, experiment tracking, and ML operations. Sub-modes — pipelines · feature-stores · experiment-tracking. Assumes Chip Huyen (ML systems author) persona. Bullets not prose.

• ml-specialist

  Use for ML model training, inference, serving, and deployment. Assumes Andrej Karpathy persona. Sub-modes — training · inference · serving. Bullets not prose.

• oracle-apex-specialist
• oracle-apexlang-specialist
• oracle-db-specialist
• oracle-fusion-specialist
• oracle-graal-specialist
• oracle-oci-specialist
• raven-approve
• raven-dashboard
• raven-debug
• raven-harden
• raven-incident
• raven-init
• raven-registry-register
• raven-registry-sync
• raven-scaffold
• raven-search
• raven-sync
• run-costs
• ui-router
• workflow-specialist

Slash commands (12)

• /raven:andie-jr

  Force-invoke Andie-jr, Raven's brownfield debug assistant, for any

• /raven:andie

  Force-invoke Andie, Raven's orchestration layer, for planning,

• /raven:raven-approve

  Use when an architect wants to approve a pending library

• /raven:raven-debug

  Use to run a full Raven boot diagnostic.

• /raven:raven-harden

  Reviews security_log.md and promotes accumulated observations into

• /raven:raven-incident

  Use to create a manual incident report for Guard events

• /raven:raven-init

  Initialize Raven for a new project. Generates manifest.json interactively, validates against schema, commits with audit trail.

• /raven:raven-registry-register

  Use to register a project in the Raven registry so it gets

• /raven:raven-registry-sync

  Use to sync all registered Raven projects to the current engine

• /raven:raven-scaffold

  Use when starting a new feature, module, or component.

• /raven:raven-search

  Use when developer wants to find, evaluate, or install a new

• /raven:raven-sync

  Use when requirements.txt or pyproject.toml has changed and you

Subagents (10)

• architecture-guard

  Use PROACTIVELY when new code files are created. During coding — advises only, never blocks. At git commit — missing architecture doc becomes a hard block via pre-commit hook after 24 hours grace period. Ensures .raven/architecture.md exists and is versioned. - Read - Bash

• claude-mem

  Use PROACTIVELY at session start and end to manage persistent memory. At start: run ide-boot.py; if load=1 Read CARD.md. At end: obsidian-log writes vault + CARD. Canonical vault is ~/RavenVault (humans). Agent boot is the card.

• db-guard

  Database discipline enforcer. Fires on PostEdit for any source file.

• guard-git-watch

  Use PROACTIVELY when any skill attempts to read sensitive files

• manifest-checker

  Use PROACTIVELY before any coding action, file creation, or tool use. Verifies .raven/manifest.json exists, is valid JSON, and has all required fields. Hard stops if manifest missing — this is the only agent that hard blocks during coding. All other agents advise only. - Read - Bash

• odoo-guard

  Odoo discipline enforcer. Spawned by odoo-specialist pre-flight and

• salesforce-guard

  Salesforce discipline enforcer. Spawned by salesforce-specialist pre-flight

• skill-guard

  Use PROACTIVELY when any skill attempts to read sensitive files

• stack-validator

  Use PROACTIVELY when any import or library is detected in code. During coding — warns and starts approval flow but does not block. At git commit — unapproved libraries are hard blocked via pre-commit hook until approval flow completes. Polars preferred over Pandas. - Read - Bash - Grep

• style-enforcer

  Use PROACTIVELY when reviewing or editing code files to advise on Giggso style standards. During coding — advise only, never block. At git commit — all style violations become hard blocks via pre-commit hook. Checks line count, type hints, docstrings, logging, naming. - Read - Bash - Grep

MCP servers (1)

• raven

  runs: python3 ${workspaceFolder}/mcp/server.py

Hooks (21)

• UserPromptSubmit

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/triage-router.py" 2>/dev/null || true

• UserPromptSubmit

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/architect-router.py" 2>/dev/null || true

• UserPromptSubmit

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/model-router.py" 2>/dev/null || true

• UserPromptSubmit

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/cve-prompt-guard.py" 2>/dev/null || true

• UserPromptSubmit

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/push-approve.py" 2>/dev/null || true

• PreToolUse

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/route-gate.py" 2>/dev/null || true

• PreToolUse

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/push-gate.py" 2>/dev/null || true

• SessionStart

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/session-start.py" 2>/dev/null || true

• SessionStart

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/ide-boot.py" 2>/dev/null || true

• SessionStart

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/model-router.py" 2>/dev/null || true

• SessionStart

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/run_costs.py" 2>/dev/null || true

• SessionStart

  rm -f "${CLAUDE_PROJECT_DIR:-.}/.raven/.push-mode" "${CLAUDE_PROJECT_DIR:-.}/.raven/.push-approved" "${CLAUDE_PROJECT_DIR:-.}/.raven/.push-notice-shown"

• PostToolUse

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/db-guard.py" 2>/dev/null || true

• PostToolUse

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/secret-scan.py" 2>/dev/null || true

• Stop

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/token-meter-write.py" 2>/dev/null || true

• Stop

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/token-guard.py" 2>/dev/null || true

• Stop

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/dashboard.py" 2>/dev/null || true

• Stop

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/obsidian-log.py" 2>/dev/null || true

• Stop

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/knowledge-extract.py" 2>/dev/null || true

• Stop

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/session-gate.py" 2>/dev/null || true

• Stop

  python3 "${CLAUDE_PLUGIN_ROOT}/scripts/code-xray.py" 2>/dev/null || true

Similar plugins