Init elektra

Skill architect-4-citadell/elektra-skills/init-elektra

Governance skills for AI coding agents. Responsible AI-driven execution — Godspeed. Built across 66+ production sessions. Zero silent failures.

Install

npx -y skills add architect-4-citadell/elektra-skills --skill init-elektra

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Run or re-run Elektra onboarding. Executes First Session Protocol (project discovery + user onboarding + memory init). Use on first install or to reset onboarding.

The file declares its own license as CC BY-NC-SA 4.0. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

2.4 KB, as published. Nobody here has run it

/init-elektra

Run or re-run Elektra's First Session Protocol.

When to use

First session in a new project (runs automatically via session-init hook)
User wants to re-onboard: changed role, new team member, different project context
Reset after cloning into a new machine or workspace

Behavior

Mode detection

Check for existing onboarding state:

if MEMORY.md exists AND contains user_profile:
  MODE = "reset"
else:
  MODE = "fresh"

Fresh mode (no prior onboarding)

Run the full First Session Protocol from CLAUDE.md:

Phase A: Self-Onboard -- read README, package.json/pyproject.toml, git history, directory structure. Save to project_discovery.md.
Phase B: Onboard User -- ask all questions in TWO batched prompts (role, experience band, team context, current work, git workflow, testing philosophy, deployment target). Save to user_profile.md.
Phase C: Initialize Memory -- create MEMORY.md with index entries.
Phase D: Configure Standing Orders -- adapt behavior based on discovered context.
Phase E: Dispatch -- classify user's current task and route to the right Standing Order.

Reset mode (re-onboarding)

Inform the user that existing onboarding data was found.
Ask: "Re-run full onboarding (project + user), or just update your user profile?"
- Full reset: archive existing project_discovery.md and user_profile.md (rename with _prev_YYYYMMDD_HHMMSS timestamp suffix), then run Phases A-E.
- Profile only: archive user_profile.md (timestamp suffix), re-run Phase B only, update MEMORY.md index.
After re-onboarding completes, confirm the new register (S1-S4) and announce readiness.

Rules

Never delete memory files outright -- archive with _prev_YYYYMMDD_HHMMSS timestamp suffix so nothing is lost.
Always run Phase A before Phase B in full reset -- project context informs which Phase B questions to skip.
Batch all onboarding questions in TWO prompts, never one-by-one.
After completing onboarding, announce: [ELEKTRA] Onboarding complete. Register: {band}. Ready.