Jev Action Check

SkillDev tools

Lets your agent screen a risky terminal command for exposed secrets, outward sends, and wide impact before it runs.

Available today. Use it from your connected AI after setup.

Add ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.

Then ask your AI: use the Jev Action Check skill

About this skill

[omh] Risky command screened by Jev: Jev action check before a risky command: secrets, outbound sends, blast radius; can only add a hold. Use when the user says: jev-action-check, jev action check, ask jev if this command is safe, jev risk check.

What this skill tells your AI

The instructions your AI receives, as published by rlaope/oh-my-hermes in skills/omh-jev-action-check/SKILL.md and read by ahel’s review.

Required: the command, working directory, and stated task. Output: hold, refuse_recommended, or no_extra_hold with its rule.

Read omh-routing/references/jev-rail.md before the first ask: consent, what leaves the machine, limits, and non-answer statuses. Prepared OMH routing is not execution or approval. Shared product, routing, compatibility, and evidence rules: omh-routing/references/skill-common-rail.md.

Contract: references/full-contract.md. Preset: references/preset.md.

Procedure

  1. Name what leaves the machine: the command, the working directory, and the stated task.
  2. Call omh_jev_ask with preset: action_check/v1 and state {command, cwd, stated_task} before anything runs.
  3. Report the outcome and rule; on hold or refuse_recommended ask the user before running, and on no_extra_hold the host's normal approval still applies.

Completion Checklist

  • Preserve workflow intent and stop conditions; load the full contract before claiming completion.
  • Reply in the user's own words and the host's own voice: OMH's record terms (surface, lane, wrapper, handoff, evidence boundary, not_observed) stay in records and tool calls, never in the sentence the user reads unless they ask about one; and when a stop condition or a decision the user owns ends the turn, offer the next action as a question rather than declaring what will not be done.
  • Report the tool's status, ask_id, served model, and cost source; a non-answer is hold with rule: not_answered:<status>, never Jev's answer.
  • Record observed delegation results; use Hermes-native subagent/delegation features when available: native subagents -> Hermes delegation when available, otherwise sequential lanes.

Recovery Notes

  • Tool missing, consent_not_observed, or refused: follow the rail's recovery section; never simulate an answer.

Workflow Lane

advisory local context

Signals

GitHub stars
3k
Forks
235
Last commit
Sep 2026
Advanced
Item type
skill
Key
omh-jev-action-check
Source
github.com/rlaope/oh-my-hermes