aliatx2017/reasonix-hermes/pre-action-gate

Trust tier 2
  • community
LLM review pending
Source check pending

Checking that the SKILL.md file is still in the repository.

Pre-write safety gate — re-read targets, lock hashes, and verify the build chain before calling any change complete.

Install

npx vskill@latest install aliatx2017/reasonix-hermes/pre-action-gate

SKILL.md

Pre-Action Gate

Load before any write, edit, or deploy. This is a _gate_ — it doesn't produce output; it blocks premature moves.

The gate

Before any file mutation — re-read the target area. Code drifts between reads; stale context is the #1 cause of garbled edits. Use content_hash on edit_file whenever you touch a file.

After any change — go build ./... must pass before you call the step done. Not "should pass," not "probably passes" — you run it and show the output. One change → one build. Iterate.

Before claiming completion — evidence must be in the _same response_ as the claim. complete_step enforces this structurally: the host rejects completions without verification commands. Don't fight it — feed it.

What this gate replaces

This skill encodes three constitution rules into one behavioral checkpoint. Load it instead of consulting the constitution directly — it's the distilled form:

  • go-vet-clean → build chain check
  • substantiate-every-claim → evidence-before-claim
  • never-say-fixed → only the user declares fixed

Failure modes

  • Stale edit — file changed since last read → garbled code. Re-read every time.
  • Premature "done" — claiming completion without build output. The host blocks it; don't waste the turn.
  • Batch-and-pray — 10 edits, then one build. One edit → one build. multi_edit for single-file batches only.

Related

  • ready-means-tested — evidence gate for status claims
  • evidence-first-reasoning — diagnostic verification under ambiguity
  • Constitution: .reasonix/constitution.json

Security

  • Security scanPassed, 100 of 100Pattern checks on the skill's files.
  • Model reviewNot runA model review is scheduled for the next rescan.

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
5

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/aliatx2017/reasonix-hermes/pre-action-gate/badge)](https://verified-skill.com/skills/aliatx2017/reasonix-hermes/pre-action-gate)

Something wrong with this skill or its page? Tell us.