aufrank
20 verified skills
defining-instrumentation
Define an instrumentation strategy that yields falsifiable evidence a feature or milestone is met. Use when asked for test/telemetry/evidence plans, validation criteria, or observability-driven success proof.
testing
agent-workflow-playbook
Guide plan → instrument → execute → validate with explicit checkpoints and questions. Use for ambiguous tasks or when enforcing a consistent agent workflow.
testing
eval-harness-kit
Build and run deterministic evaluation suites for agent workflows (single-turn or agentic). Use when you need reproducible eval runs with manifests, graders, metrics, and JSONL logs for capability or regression tracking.
development
long-running-harness-kit
Design long-running agent harnesses with resumable checkpoints and initializer/coder handoffs. Use when tasks span multiple sessions or require recovery after interruption.
development
querying-jira-programmatically
Run Jira MCP searches (Rovo or JQL) via mcpc-backed Python wrappers with cached tool schemas and file-backed outputs. Use when Codex needs deterministic Jira queries using @jira and mcpc (no direct HTTP).
tools
querying-notion-programmatically
Run Notion MCP searches for pages/blocks and fetch page comments via mcpc-backed Python wrappers. Use when Codex needs deterministic, file-backed Notion queries (filters come later).
tools
skills-audit
Audit skill folders for guideline compliance and portability. Use when validating SKILL.md files against skill_development_guidelines.md or checking for non-portable examples, hard-coded paths, or missing references.
development
slicing-long-contexts
Use when a prompt or corpus is long/dense (multi-docs, logs, codebases) and you want a reproducible map/reduce pipeline. Trigger this skill to slice inputs, run per-slice codex/gemini subcalls, and aggregate results with manifests/logs via the slice runner (preferred default), falling back to manual REPL slicing only if needed.
development
using-mcp-tools-with-mcpc
Use mcpc CLI to interact with MCP servers - call tools, read resources, get prompts. Use when working with Model Context Protocol servers, calling MCP tools, or accessing MCP resources programmatically; prefer key:=value bindings over raw JSON bodies.
tools
workflow-or-agent-decider
Decide between a scripted workflow and an autonomous agent harness, then scaffold the chosen path. Use when scoping new agentic systems or tool integrations.
tools
pincer-planning
Bidirectional planning technique that pairs top-down framing (goals, constraints, success criteria) with bottom-up inventory (facts, assets, risks, unknowns), then stress-tests both sides with targeted questions before converging into a gap-checked plan. Use when: - A task is ambiguous and top-down reasoning stalls because key unknowns are unresolved - There is a concrete inventory of facts/assets but no clear unifying goal or structure - An existing plan looks complete but may have hidden assumptions or blind spots - A design, architecture, or strategy needs to be stress-tested before execution - The user asks to "plan this out", "think through this", "help me figure out how to approach X", or "what am I missing" — especially on fuzzy, multi-step, or design-heavy tasks - A plan has too many gaps and needs convergence before implementation begins Do NOT trigger for well-specified tasks where the path forward is already clear.
development
austin-frank-voice
Write in Austin Frank's voice and style. Use this skill whenever generating text that should sound like Austin — strategy docs, charters, proposals, business cases, vision documents, staffing requests, stakeholder updates, cover letters, mission statements, org design documents, or any professional prose where the user wants Austin's distinctive voice. Also use when the user asks to review, edit, or improve a draft for voice consistency, or when they reference "my style", "my voice", "write like me", or "Austin's style".
documentation
tool-quality-audit
Audit tool integrations for deterministic behavior, error contracts, and logging before agents depend on them. Use when adding or updating tools or MCP servers.
tools
creating-skills
General skill authoring and improvement. Use when creating or updating any skill (with or without MCP), selecting structure, generating scaffolds, or packaging resources (scripts/references/assets/templates) that follow progressive disclosure, deterministic execution, and restartable artifacts.
tools
creating-mcp-code-mode-skills
A meta-skill for authoring high-performance, verifiable, long-running MCP skills using Code Mode. This skill blends Anthropic and OpenAI skill-authoring guidance with Code-Mode-first, MCP-backed execution, dynamic context discovery, and file-backed agent harnesses.
tools
parallelizing-with-worktrees
Create and manage Git worktrees to parallelize agent tasks safely. Use when you need multiple concurrent checkouts (sub-agents, parallel edits, or isolated experiments) and want deterministic setup/cleanup with manifests and logs.
testing
reporting-situation
Generates a comprehensive situation report by tracking collaborators, topics, and teams across Google Workspace, Notion, and Jira. Maintains a persistent knowledge graph and item tracker.
development
working-with-notion-programmatically
Use mcpc to interact with the Notion MCP server: connect sessions, search workspace content, fetch pages/databases, and run helper scripts for common Notion actions.
tools
inspecting-the-environment
Fast environment briefing for agents and subagents. Use at session start to learn OS/shell, container/WSL/VM status, git repo + upstream + dirty state, Python venv status/locations, markdown folders to read, and availability of common tools (uv, mcpc, rg/grep/jq/git/python/node/npm/pnpm, etc.).
tools
running-dag-pipelines
Build and execute modular DAG workflows for long-context processing using slice/map/reduce/recurse/compact/filter operators. Use for one-shot batch jobs, standalone map-reduce pipelines, or when the context-dag plugin is not installed. Trigger when input exceeds the model's context window, when reproducible logged pipelines are needed, or when multi-level recursive processing is required. If context-dag is installed, the plugin's bundled dag_runner.py provides the same capability with persistent artifact storage.
tools