patientvibes-skills

User-global slash commands for the chorus-dev VM environment — /check (gate pipeline), /db (docker-exec psql/sqlplus against the 4 chorus DBs), /oc (delegate to opencode). Tightly coupled to chorus-dev's specific tooling; NOT portable to other machines.

3 commands

AI-driven PR review — `pr-review` skill for user-invoked review of the current branch, `pr-review` subagent for programmatic dispatch by other agents (e.g. ship). Both wrap the agent-tool-pr-reviewer CLI.

1 skill · 1 agent

Gemini CLI as a second-opinion / large-context research assistant for the current repo. Spawns `gemini -p` for whole-codebase reasoning that doesn't fit in Claude's working context.

1 skill

Convention for loading API keys + secrets in agent-tool-* / agent-harness-* CLIs. Tool reads ~/.config/<tool-name>/env (mode 600) as a fallback when the env var isn't set. Pairs cleanly with 1Password `op item get` workflows.

1 skill

Iterative Claude ↔ Gemini collaborative planning loop. Claude drafts, Gemini critiques, Claude revises — repeat until convergence.

1 skill

Skill for analyzing SS&C AWD CSD binary forms with a LangGraph ReAct agent (paired with agent-harness-chorus-csd-analyzer)

1 skill

Find and fix bare `except:` clauses and overly broad `except Exception` catches across a Python codebase. Narrows to specific exception types (ImportError, ConnectionError, ValueError, pydantic.ValidationError, etc.) and adds logging where exceptions were silently swallowed. Leaves test files alone.

1 skill

Turn any folder of files (code, docs, papers, images, videos) into a navigable knowledge graph with community detection and audit trail. Outputs interactive HTML, GraphRAG-ready JSON, and a plain-language report.

1 skill

Pattern: gate output by confidence, route low-confidence items to a queue, feed corrections back as wiki rules. Used by card-extractor; pairs with agent-tool-review-queue + agent-tool-knowledge-wiki.

1 skill

Subagent for repo documentation cleanup and AI-agent instruction consolidation (AGENTS.md / CLAUDE.md / GEMINI.md / Copilot). Nine-phase workflow that scales to the task. Removes stale artifacts without changing application behavior. Output is a PR, not a merge.

1 agent

Subagent driving an approved change from branch → PR → external review → merge. External reviewer: GLM 5.3 via a read-only opencode `review` agent (repo access); falls back to the `pr-review` subagent pinned to GLM 5.3, then /code-review. Triggered only after the user has approved a plan and authorized full automation through merge.

1 agent

Pattern: extract → verify → retry-once-with-feedback. Used by chorus-csd-analyzer and card-extractor harnesses.

1 skill

Six production-tuned VLM prompts for scanned-document extraction (paired with agent-harness-card-extractor)

1 skill