Skip to content

rlaope/oh-my-hermes / skills/omh-routing

omh-routing

Verified

[omh] Choosing among OMH skills for a request: router guidance for using oh-my-hermes workflow skills inside Hermes Agent.

references/evidence-boundaries.md

View on GitHub
# OMH Evidence Boundaries

OMH is a Hermes-native wrapper orchestration layer. It keeps Hermes responsible for chat intake, clarification, source-backed research, planning, and status narration while coding-heavy work is prepared as explicit handoff and tracked only when observed.

## Prepared Versus Observed

Prepared routing, plans, task cards, and handoffs are not execution evidence. `prepared_not_observed` is not implementation, review, CI, merge-readiness, merge evidence, plugin runtime use, or proof that another agent acted.

## Runtime Evidence

When local shell access or a bot wrapper is available, record prepared handoffs and observed workflow evidence under `.omh/runtime/`:

```sh
omh coding delegate --source discord --executor <owner> --record "risky refactor"
omh runtime delegate --run <run-id> --requested --not-observed --result not_observed
```

Record only what is observed. If Hermes or a chosen oh-my runtime does not expose delegation metadata, use `not_observed` or `not_available` instead of implying a specialist lane ran.

## Multi-Agent Target Awareness

Wrappers may report `omh_target_topology/v1` when a workspace moves between one Hermes agent target and multiple Hermes agent targets. Treat that topology as setup evidence only. If `active_agent_count` is greater than one, bind this workflow to the current target and thread, name the target boundary in status, and do not claim another Hermes agent observed, accepted, or executed the workflow unless target-specific evidence exists.

If a wrapper reports `single_to_multi` or `multi_to_single`, answer with one concise target-change comment. If the wrapper exposes an `apply_target_change` action and the user accepts it, persist the target registry update; otherwise keep the workflow scoped to the current thread target and ask before assuming multi-agent behavior. A skill that does not need multiple agents should continue as a single-target workflow even when multiple targets are known.

## Memory Context

`memory_review_card/v1` is separate from `status_card/v1`; `handoff_context_pack/v1` may be attached to executor handoffs only when unresolved conflicts are absent.

## Goal Status

`goal_status_card/v1` and `goal_continuation/v1` are goal-execution payloads separate from generic `status_card/v1`; they must name the next action instead of merely summarizing work.

## Hermes Compatibility

- Use Hermes tools and subagents when available.
- Replace unavailable goal tools with file-backed checklists or ledgers.
- Replace unavailable question renderers with one direct question through the current Hermes surface.
- Keep shell bridge behavior explicit and opt-in.
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall