rlaope

[omh] Large multi-part job to run on its own: plan it as a graph of small nodes, run the ready ones in parallel through Hermes' delegate_task, verify the goal with a separate node, and repair failed nodes before reporting. Use when the user says: dag-ulw, dag…

[omh] Technical proposal facing adversarial scrutiny: independent perspectives attack a proposal, then distill into a bundle a separate planner consumes. Use when the user says: adversarial-consensus, adversarial planning, adversarial plan review, red team…

[omh] AI agent progress for managers: help managers inspect AI-agent progress, blockers, quality gates, and throughput levers. Use when the user says: agent-ops-review, agent ops review, agent productivity, operator productivity, manager view, quality…

[omh] Career search to interview: find verified opportunities, tailor evidence-grounded applications and track confirmed outcomes. Use when the user says: ai-job-search, job search, evaluate this job posting, tailor my resume, tailor my cv, prepare for my job…

[omh] Messy or AI-generated code to clean up: delete AI-generated slop, dead code, and duplication while observable behavior stays identical. Use when the user says: ai-slop-cleaner, cleanup, deslop, refactor, risky, behavior-preserving refactor, risk…

[omh] Attack paths into an operated system: turn a system's components and data flows into assets, trust boundaries, attack scenarios, controls, and the security test that proves each control holds. Use when the user says: application-threat-model,…

[omh] Outside AI critique wanted: consulting an external advisor when configured. Use when the user says: ask, external advisor, ask claude, ask gemini, consult claude, consult gemini, opinion from claude, opinion from gemini.

[omh] Public variant claims to cross-check: retrieve and reconcile source-labelled biomedical evidence with identity, conflict and freshness checks. Use when the user says: biomedical-evidence, biomedical evidence reconciliation, ClinVar cross-check, variant…

Evidence-driven causal tracing — competing hypotheses, falsification, rebuttal round, recommended next probes

Feature validation gate — research, define purpose, stress-test from real user perspective before building

Maximum parallelism burst — all tasks execute simultaneously with no coordination overhead

Persistent completion mode — tasks must finish fully with verify/fix loops until done

Queue and auto-process multiple GitHub issues — each gets its own branch, team, PR, and merge

Execute tasks in parallel with dynamically assigned agents per task

Ask targeted clarifying questions to uncover missed requirements before execution

1000-iteration thought simulation — plan, build, validate, critique, amplify

Invoke when deciding what to load into context, what to persist across turns or sessions, and what to drop — reads only what the task needs, writes durable handoffs, avoids context pollution, and chooses when to re-derive versus recall.

Invoke when deciding whether to split work across parallel agents/sessions or do it inline — tests scope independence, weighs coordination cost against the parallelism win, and sets an observation cadence for delegated work.

Build an accurate mental model of an unfamiliar codebase or problem before making any edit — structural scan, entry-point location, one end-to-end trace, and an explicit known-vs-assumed ledger.

Systematically enumerate how code can break — boundaries, concurrency, partial failure, malicious input, time, resource exhaustion — before writing defenses, then convert the list into tests and guards while pruning paranoia that does not pay rent.

Invoke when writing any status, completion, or result report — calibrates claims to the evidence actually gathered, marks the unmeasured as open, reports failures plainly, and distinguishes "verified at layer X" from "should work".

Debug by maintaining 2-4 competing hypotheses with explicit evidence and kill criteria, choosing each probe to maximally discriminate between them instead of confirming a favorite.

Decompose a large or risky change into a sequence of independently safe intermediate states, so the system stays green at every step and any step can be reverted alone.

Invoke when an implementation attempt produces a failure signal (test failure, compile error, wrong output, crash, perf regression) and you must decide whether to retry, rethink, or back out — instead of hammering the same fix.

Aura-field auto-routing — read an incoming request's problem shape and engage the right nen abilities, stacked in dependency order when the work spans more than one discipline. Use when the user says "nen", "use nen", "nen mode", "run this through nen",…

Skill forging — distilling a team's recurring, hard-won practice into a new installable SKILL.md that meets the nen contract. Use when the user says "write a skill", "create a skill for", "turn this workflow into a skill", "codify our process", or "SKILL.md";…

Orchestration engineering for multi-agent and deep-research work — decomposition, routing, prompt design, and verified delegation. Use when the user says "plan this", "break this down", "delegate", "orchestrate", "subagent", "task distribution", "parallel…

Pre-commitment engineering — declare falsifiable vows before the work starts and audit every one before claiming done, discipline pledges distilled from the engaged skills' Done means plus scope pledges bounding the blast radius. Use when the user says "don't…

Failure-mode-driven reliability engineering — enumerate how it breaks, give every mode a verdict, prove every handler with a reproducing test. Use when the user says "flaky", "harden this before launch", "what if this fails", "error handling", "timeout",…

Reverse engineering and mechanism-comparison research — when the docs run out, read the artifact itself and prove how the thing actually works. Use when the user asks "how does <tool> actually do it", says "reverse engineer this", "undocumented behavior",…

Behavior-preserving refactoring under characterization tests, plus severity-rated code review. Use when the user says "refactor", "clean up", "tech debt", "code review", "extract", "rename", "restructure", "this file is a mess", or "impossible to modify";…

Launch, positioning, and promotion — making finished work land with the audience that should care. Use when the user says "launch", "announce", "release notes", "changelog", "blog post", "tweet thread", "X thread", "Show HN", "Hacker News post", "how do I…

Audit and update Argus external distribution surfaces (Spring Boot starter, Helm chart, Docker, install scripts, Homebrew Formula, SDKMAN, GitHub Actions, action.yml) against gradle.properties and known-good upstream versions. Detects version drift, floating…

Audit and update Argus user-facing docs (README, docs/**/*.md, site/*.html) against the code truth. Detects version drift, command-count drift, i18n parity violations, and stale CLI examples. Use before a release, or any time the user asks to "sync docs",…

Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code in this repo to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.

Continuous JVM monitoring + optimization + troubleshooting harness on top of the Argus CLI. Use when the user wants to watch a running JVM for trouble (GC regression, heap leak, thread growth, pause-time spikes) over a window of time and get a single…

Run the recurring repo delivery loop: choose or clarify the next goal, run the real ralplan gate, implement through ultragoal or ultrawork, open a PR, run code review, wait for CI/DCO, fix failures, then ask for human-in-the-loop merge approval before merging…