AI Engineering

Agent Incident Postmortem runs a blameless postmortem for incidents caused by AI agents or LLM features, such as hallucinations, prompt injection, or runaway tool use. Use it to reconstruct traces, analyze root causes across layers, and produce corrective actions including a permanent regression case.

Agent Observability specifies the tracing, metrics, and alerts for an AI agent or LLM feature in production. Use it when you need to log LLM app behavior, design agent spans, define quality and cost monitors, or determine if an agent is misbehaving.

Agent Readiness Audit evaluates whether AI agents can actually use a product by auditing docs, APIs, onboarding, errors, and discoverability from a non-human perspective. Use it when preparing for agentic traffic or when agents fail against your site or API, and get a scored readiness report with prioritized fixes.

Agent Spec defines the goal, tools, control loop, guardrails, memory, escalation, evaluation, and failure handling for an autonomous or tool-using AI agent before it is built. Use it when asked to design an AI agent, define its tools and permissions, or write an agent spec/PRD.

Takt orchestrates sub-agents through YAML-defined workflows using codex exec, enabling multi-agent collaboration with steps, rules, and loop monitors. Use it to delegate coding, review, and design tasks to AI sub-agents in a controlled sequence.

Takt orchestrates a team of AI agents according to a YAML workflow definition. Use it to run multi-step, multi-agent workflows where a team lead delegates tasks, evaluates rules, and manages reports.

Helmor CLI remote-controls Helmor from the terminal, letting you inspect data/settings, manage repos, workspaces, sessions and files, send prompts to agents, use GitHub integration, inspect scripts, run as an MCP server, and more. Use it for CLI workflows including stacked PRs and restacking.

Autocontext is a control plane for evaluating agent behavior, preserving run artifacts, exporting training data, and distilling stable behavior into local runtimes. Use it when a Hermes agent needs to run scenarios, inspect Hermes Curator state, or prepare local MLX/CUDA training data through the autoctx CLI.

Autocontext Consumer reads and moves knowledge produced by Autocontext, including playbooks, lessons, and hints. Use it to check what has been learned, read current playbooks, and export or import knowledge between checkouts.

Autocontext Creator runs an Autocontext improvement loop over a task and writes what it learned to disk. Use it when an agent needs to create knowledge through scenarios, judge or improve a single output, or inspect what a run produced.

Grid CTF Ops provides operational knowledge for the grid_ctf scenario, including strategy playbook, lessons learned, and resource references. Use when generating, evaluating, coaching, or debugging grid_ctf strategies.

AI-Assisted Prototyping helps product leaders build functional prototypes from natural language or visual mocks using AI coding tools. Use it to validate ideas, bypass engineering bottlenecks, and hand off high-fidelity reference code.