Skip to content
Back to skills

Soul Audit

ASecurity

6-phase interactive interview that generates the agent's identity (identity/soul), operator profile (identity/user), access posture (identity/access-policy), and operational cadence (identity/heartbeat) as brain pages. Re-runnable anytime to update any section.

  • 11 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added September 28, 2026
ai-agentsgo

Works with

  • mcp

Security analysis

A100/100

Scanned September 28, 2026

npx -y skills add timurgaleev/memex --skill soul-audit --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Soul Audit?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Soul Audit
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/timurgaleev-soul-audit/badge)](https://www.skillsdirectory.com/skills/timurgaleev-soul-audit)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

SKILL.md
---
name: soul-audit
version: 1.0.0
description: |
  6-phase interactive interview that generates the agent's identity
  (identity/soul), operator profile (identity/user), access posture
  (identity/access-policy), and operational cadence (identity/heartbeat)
  as brain pages. Re-runnable anytime to update any section.
triggers:
  - "soul audit"
  - "customize agent"
  - "who am I"
  - "set up identity"
  - "change my agent's personality"
tools:
  - page_put
  - page_get
  - get_brain_identity
  - whoami
mutating: true
writes_pages: true
writes_to:
  - identity/
---

# Soul Audit — Agent Identity Builder

Generate the agent's identity and operational configuration through an interactive
interview. Each phase produces a brain page. Any phase can be re-run independently
to update. The `get_brain_identity` tool serves these pages back to any connected
agent, so what you write here shapes every future session.

**IMPORTANT:** This skill generates content from the OPERATOR'S OWN ANSWERS. It
NEVER ships pre-filled content. The minimal scaffolds in Default Mode are
skeletons, not defaults.

## Contract

This skill guarantees:
- `identity/soul` generated from the operator's description of agent identity, vibe, mission
- `identity/user` generated from the operator's self-description (role, projects, key people)
- `identity/access-policy` generated with an explicit access posture
- `identity/heartbeat` generated with the operational cadence the operator chooses
- Each phase is independent and re-runnable
- Default mode (skip soul-audit): writes minimal scaffold pages via `page_put`

## Phases

### Phase 1: Identity Interview
Ask: "What is this agent to you? Research partner? Executive assistant? Thinking partner? All of the above?"
Generate: `identity/soul` identity section (`page_put`).

### Phase 2: Vibe Calibration
Show 3-4 communication style examples:
- **Formal:** "I've prepared a comprehensive analysis of the situation..."
- **Direct:** "Here's what's happening. Three things matter."
- **Technical:** "The root cause is in the connection pooling. Here's the fix."
- **Casual:** "Yeah so basically the thing is broken because X. Easy fix."
Ask which feels right. Generate: `identity/soul` vibe + communication style sections.

### Phase 3: Mission Mapping
Ask: "What are your top 3-5 goals? What are you trying to accomplish?"
Generate: `identity/soul` mission + operating principles sections.

### Phase 4: Operator Profile
Ask: "Tell me about yourself. What do you do? What are you working on? Who are the key people in your world?"
Generate: `identity/user` with role, projects, key people, communication preferences.

### Phase 5: Boundaries
This is a one-operator brain, but it has two ingress surfaces: the internal MCP
surface (full tool roster) and the public MCP surface (redacted reads plus a
constrained write set). Ask: "What should the public surface be allowed to see
or write? Anything that must stay internal-only?"
Generate: `identity/access-policy` documenting the posture — what each surface
is for, what never leaves the internal side, and any content classes agents
should refuse to surface publicly. Note: enforcement lives in the server's
ingress config; this page is the human-readable policy agents follow.

### Phase 6: Operational Cadence
Ask: "How often should the agent check in? Morning briefing? End of day summary? What recurring work do you want?"
Generate: `identity/heartbeat` with the cadence. Recurring server-side
maintenance is already covered by the brain's background cycle; extra recurring
work is scheduled operator-side (systemd timers on the host) or via the agent
harness's own scheduler — record the chosen cadence on the page either way.

## Default Mode (Skip Soul-Audit)

If the operator skips soul-audit on first boot:
- Write `identity/soul` as a minimal scaffold ("knowledge-first agent with persistent memory")
- Write `identity/user` auto-populated with the operator identity from `whoami`
- Write `identity/access-policy` as owner-only / internal-surface-first
- Write `identity/heartbeat` with the default cadence (rely on the brain's cycle)

## Output Format

Four brain pages generated/updated. Report: "Soul audit complete: identity/soul,
identity/user, identity/access-policy, identity/heartbeat written. Re-run any
phase anytime to update."

## Anti-Patterns

- Shipping pre-filled soul or user content (privacy violation)
- Making soul-audit mandatory on first boot (high friction, optional is better)
- Asking all 6 phases in one go (overwhelming, each is independent)
- Not offering to re-run individual phases

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.