ham

Review executor prompts for red flags, alignment issues, blockers, scope drift, and evidence gaps before any non-trivial execution.

Structured diagnosis loop: reproduce, minimize, hypothesize, instrument, fix, and regression-test.

Grilling session for plan alignment against domain language and ADRs, with inline doc updates when decisions are finalized.

Summarize the current thread into a compact handoff artifact for a fresh agent session.

PARKED/MANUAL ONLY: Identify architecture deepening opportunities and refactor candidates.

Repo-local setup for Matt Pocock engineering skills in HAM. Configures issue tracker, triage labels, and domain doc locations without touching runtime or deploy settings.

Red-green-refactor loop using vertical slices and behavior-focused tests through public interfaces.

Convert accepted plans/specs into independent vertical-slice GitHub issues with explicit dependencies.

Turn current conversation context into a PRD-style implementation issue with explicit module and testing decisions.

PARKED/MANUAL ONLY: Triage issue workflow through category/state labels when explicitly requested.

Explain a subsystem at a higher abstraction level and map related modules/callers before editing.

Minimal starter runbook for Cloud agents to install, run, and test Ham quickly. Covers backend/frontend startup, chat gateway modes, Cursor key setup, browser runtime checks, and how to keep this skill current as new runbook knowledge is found.

How Ham previews and launches allowlisted Factory droid exec workflows from chat (readonly audit vs low-risk edit), runner assumptions, and hard security limits. Instructional only — policy lives in code.

Wire memory_heist ContextBuilder into Hermes-supervised role prompts in swarm_agency.py using a single shared ProjectContext and per-role render budgets. Use when connecting Hermes-led flows to repo context, adjusting budgets, or integrating SessionMemory. (No CrewAI.)

Guides conversational navigation of the Ham dashboard: settings sections, projects, Droids/registry, runs and activity, and how to scaffold sub-agent workflows or new assets using existing APIs and configs. Use when the user wants natural-language help finding settings, understanding Hermes vs Droid roles, creating workflows or agents, or pairing chat with the workspace UI without reading raw logs.

Validate the Hermes supervisory critic review loop: verify Hermes receives correct context, invokes HermesReviewer.evaluate(), and preserves learning signals for later persistence. Use when modifying hermes_feedback.py, the supervisory review path, or the review pipeline.

Audit token budgets across the Ham swarm to prevent context window overflows. Use when checking prompt sizes, reviewing MAX_* constants, or diagnosing truncation issues in agent outputs.

Harden memory_heist.py against token blowouts, unbounded growth, cross-platform path bugs, stale Claude references, and continuation/parser marker drift. Use when modifying memory_heist.py, fixing compaction logic, capping diffs, or adding safety limits to the Context Engine.

Test that memory_heist.py correctly scans the repo, discovers configs and instructions, captures git state, and compacts sessions without data loss. Use when adding tests for Context Engine changes or verifying cross-platform behavior.