### Manual kitty install steps

Source: https://docs.ourlifeos.ai/Terminal__Helm

Manual install steps: install kitty and a Nerd Font, copy the config set into ~/.config/kitty (overwrites existing), then launch kitty. Does not create the Helm.app wrapper or helm command. Helper scripts reference ~/.config/kitty by absolute path, so install there.

```bash
# 1. Kitty + a Nerd Font.  macOS:
brew install --cask kitty font-hack-nerd-font
#    Linux: install kitty from your package manager or https://sw.kovidgoyal.net/kitty/
#    and install any Nerd Font (https://www.nerdfonts.com) for the tab-bar icons.

# 2. Copy the config set into place (this overwrites — back up ~/.config/kitty first)
mkdir -p ~/.config/kitty
cp -R ~/.claude/LIFEOS/DOCUMENTATION/Terminal/kitty/ ~/.config/kitty/

# 3. Launch kitty (or reload with shift+cmd+r inside it)
```

--------------------------------

### Recovery and setup commands for LifeOS work tracking

Source: https://docs.ourlifeos.ai/Work__WorkSystem

Commands for recovery and setup: re-authenticate GitHub CLI, seed labels, restart Pulse, and install the work sweep launchd job.

```bash
gh auth login
```

```bash
bun BootstrapLabels.ts
```

```bash
bash ~/.claude/LIFEOS/PULSE/manage.sh restart
```

```bash
bun ~/.claude/LIFEOS/TOOLS/InstallWorkSweep.ts
```

```bash
bun ~/.claude/LIFEOS/TOOLS/InstallWorkSweep.ts
```

```bash
bun ~/.claude/LIFEOS/TOOLS/InstallWorkSweep.ts
```

--------------------------------

### Example 2: Publish

Source: https://docs.ourlifeos.ai/Skills__SkillSystem

Example workflow for publishing blog content, showing the user request and the steps taken.

```markdown
User: "Publish the AI agents post"
→ Invokes Publish workflow
→ Runs build validation
→ Deploys to Cloudflare Pages
```

--------------------------------

### Install UV package manager

Source: https://docs.ourlifeos.ai/Tools__Tools

Installs the UV package manager, a prerequisite for running the script.

```bash
curl -LsSf https://astral.sh/uv/install.sh | sh
```

--------------------------------

### Run Helm install script

Source: https://docs.ourlifeos.ai/Terminal__Helm

Preferred one-command install. Installs kitty and a Nerd Font if missing, backs up an existing ~/.config/kitty directory with a timestamp, links Helm's config, builds a Helm.app wrapper, and adds helm to PATH. On Linux, expects kitty to be pre-installed.

```bash
bash ~/.claude/LIFEOS/DOCUMENTATION/Terminal/install.sh
```

--------------------------------

### Install Blog discovery

Source: https://docs.ourlifeos.ai/Services__BackgroundServices

Installs the Blog discovery service (com.lifeos.blogdiscovery). This is an opt-in service that runs daily/scheduled.

```bash
bun ~/.claude/LIFEOS/TOOLS/InstallBlogDiscovery.ts
```

--------------------------------

### Example 1: Write new content

Source: https://docs.ourlifeos.ai/Skills__SkillSystem

Example workflow for writing new blog content, showing the user request and the steps taken.

```markdown
User: "Write a post about AI agents for the blog"
→ Invokes Create workflow
→ Drafts content in scratchpad/
→ Opens dev server preview at localhost:5173
```

--------------------------------

### Command Structure Example

Source: https://docs.ourlifeos.ai/Tools__CliFirstArchitecture

Shows a hierarchical command structure with subcommands and flags, using 'evals' as an example.

```bash
# Good: Hierarchical, clear structure
tool command subcommand --flag value

# Examples:
evals use-case create --name foo
evals test-case add --use-case foo --file test.json
evals run --use-case foo --model claude-3-5-sonnet
```

--------------------------------

### Install Derived-file sync

Source: https://docs.ourlifeos.ai/Services__BackgroundServices

Installs the Derived-file sync service (com.lifeos.derivedsync). This is an opt-in service that runs on file-change.

```bash
bun ~/.claude/LIFEOS/TOOLS/InstallDerivedSync.ts
```

--------------------------------

### Well-formed ISA example for adding --json flag

Source: https://docs.ourlifeos.ai/ISA__ISAFormat

A complete ISA file example that satisfies the format contract for a small CLI tool. It includes frontmatter, goal, claims, and test strategy. The example shows minimal frontmatter, sequential claim numbering, and empty sections omitted.

```markdown
---
task: "add --json output flag to the CLI"
slug: 20260723-140000_cli-json-flag
phase: complete
progress: 3/3
started: 2026-07-23T14:00:00Z
updated: 2026-07-23T14:40:00Z
---

## Goal
The CLI accepts a `--json` flag that prints one valid JSON object carrying the
same data it prints as text, and the default text output is unchanged.

## Claims
- [x] C1: `--json` prints output that parses as a single JSON object.
- [x] C2: The JSON carries the same fields the text output shows.
- [x] C3: Anti: running without `--json` produces byte-identical text to before.

## Test Strategy
| claim | type     | check                                        | threshold | tool | anchors_to |
| C1    | bash     | `cli --json \| jq .` exits 0                 | exit 0    | bash | literal |
| C2    | bun-test | fields(json) == fields(text)                 | equal     | bun test test/cli.test.ts -t "json parity" | literal |
| C3    | bash     | `cli` output diff vs saved baseline is empty | empty     | bash | derived: text-parity |
```

--------------------------------

### Install Pulse Menu Bar App

Source: https://docs.ourlifeos.ai/Pulse__PulseSystem

Builds and deploys the menu bar app to ~/Applications and installs the launchd plist. Run from the MenuBar directory.

```bash
cd ~/.claude/LIFEOS/PULSE/MenuBar
bash install.sh    # Builds, deploys to ~/Applications, installs plist
```

--------------------------------

### Build and install LifeOS Pulse menu bar app

Source: https://docs.ourlifeos.ai/docs/Pulse__PulseSystem

Commands to build and install the menu bar app. Run from the MenuBar directory: first execute install.sh to compile, deploy to ~/Applications, and install the plist; then use build.sh to recompile after code changes.

```bash
cd ~/.claude/LIFEOS/PULSE/MenuBar
bash install.sh    # Builds, deploys to ~/Applications, installs plist
```

```bash
bash build.sh      # Compiles PulseMenuBar.swift → LifeOS Pulse.app
```

--------------------------------

### Always Use CLI - Good vs Bad Example

Source: https://docs.ourlifeos.ai/Tools__CliFirstArchitecture

Shows the recommended approach of using the CLI directly versus replicating its functionality manually. The good example uses `bash('evals run --use-case newsletter-summary')`, while the bad example attempts to manually load config and test cases.

```javascript
// Good: Use the CLI
await bash('evals run --use-case newsletter-summary');

// Bad: Replicate CLI functionality
const config = await readYaml('use-cases/newsletter-summary/config.yaml');
const tests = await loadTestCases(config);
for (const test of tests) {
  // ... manual implementation
}
```

--------------------------------

### Install and Configure Hermes

Source: https://docs.ourlifeos.ai/docs/Hermes__HermesSidecar

Commands to install Hermes, authenticate, and mount the LifeOS environment.

```bash
# 1. Install Hermes (pinned, no Playwright — LifeOS uses Interceptor)
curl -fsSL https://hermes-agent.nousresearch.com/install.sh -o install.sh
# review it, then:
bash install.sh --non-interactive --skip-setup --skip-browser --commit <reviewed-sha>

# 2. Authenticate (its OWN session — never import a sibling agent's tokens)
hermes auth add openai-codex --type oauth --no-browser

# 3. Mount LifeOS
bun LIFEOS/HERMES/Mount.ts

# 4. Install the launcher (below) and start from it, never from bare `hermes`
```

--------------------------------

### Install Usage aggregator

Source: https://docs.ourlifeos.ai/Services__BackgroundServices

Installs the Usage aggregator service (com.lifeos.usage-aggregator). This is an opt-in service that runs daily/scheduled.

```bash
bun ~/.claude/LIFEOS/TOOLS/InstallUsageAggregator.ts
```

--------------------------------

### LifeOS Services CLI commands

Source: https://docs.ourlifeos.ai/Services__BackgroundServices

Commands to check status, install all or specific services, uninstall a service, and regenerate documentation. Install and uninstall require launchctl privileges; install without --yes is a dry preview.

```bash
bun ~/.claude/LIFEOS/TOOLS/Services.ts status              # what's running vs installed vs available
bun ~/.claude/LIFEOS/TOOLS/Services.ts install --all --yes # stand up every service (macOS)
bun ~/.claude/LIFEOS/TOOLS/Services.ts install --only worksweep,amberroute --yes
bun ~/.claude/LIFEOS/TOOLS/Services.ts uninstall --only amberroute
bun ~/.claude/LIFEOS/TOOLS/Services.ts doc                 # regenerate the table below
```

--------------------------------

### Agent model selection examples

Source: https://docs.ourlifeos.ai/Delegation__DelegationSystem

Examples showing correct and incorrect usage of the Agent tool's model parameter. The first example is marked WRONG because it defaults to Opus, which is slow for simple checks. The subsequent examples show the RIGHT model for different task types: haiku for simple visual checks, sonnet for standard coding, and opus for complex architectural planning.

```javascript
// WRONG - defaults to Opus, takes minutes
Agent({ prompt: "Check if blue bar exists on website", subagent_type: "general-purpose" })

// RIGHT - Haiku for simple visual check
Agent({ prompt: "Check if blue bar exists on website", subagent_type: "general-purpose", model: "haiku" })

// RIGHT - Sonnet for standard coding task
Agent({ prompt: "Senior engineer, TDD. Implement the login form validation", subagent_type: "general-purpose", model: "sonnet" })

// RIGHT - Opus for complex architectural planning
Agent({ prompt: "System design / distributed systems. Design the distributed caching strategy", subagent_type: "general-purpose", model: "opus" })
```

--------------------------------

### Conversational task management examples

Source: https://docs.ourlifeos.ai/Pulse__DaSubsystem

Example user and DA interactions for listing and cancelling tasks.

```text
User: "what reminders do I have?"
DA: Lists active scheduled tasks

User: "cancel the PR review reminder"
DA: Finds matching task, sets status to cancelled
```

--------------------------------

### Install Health sync

Source: https://docs.ourlifeos.ai/Services__BackgroundServices

Installs the Health sync service (com.lifeos.healthsync). This is an opt-in service that runs every hour.

```bash
bun ~/.claude/LIFEOS/TOOLS/InstallHealthSync.ts
```

--------------------------------

### Skill Example 2: Another Use Case

Source: https://docs.ourlifeos.ai/Skills__SkillSystem

Template for another skill example showing a user request, the process, and the output.

```markdown
User: "[Another typical request]"
→ [Process]
→ [Output]
```

--------------------------------

### Minimal SKILL.md example

Source: https://docs.ourlifeos.ai/Skills__SkillSystem

Example of a minimal SKILL.md file for the Art skill. Includes YAML frontmatter, workflow routing table, quick reference, and pointers to detailed docs via SkillSearch.

```markdown
---
name: Art
description: Visual content system. USE WHEN art, header images, visualizations, diagrams.
---

# Art Skill

Complete visual content system using **charcoal architectural sketch** aesthetic.

## Workflow Routing

| Trigger | Workflow |
|---------|----------|
| Blog header/editorial | `Workflows/Essay.md` |
| Technical diagram | `Workflows/TechnicalDiagrams.md` |
| Mermaid flowchart | `Workflows/Mermaid.md` |

## Quick Reference

**Aesthetic:** Charcoal architectural sketch
**Model:** nano-banana-pro
**Output:** Always $LIFEOS_DOWNLOADS_DIR (default ~/Downloads/ when unset) first

**Full Documentation:**
- Aesthetic guide: `SkillSearch('art aesthetic')` → loads Aesthetic.md
- Examples: `SkillSearch('art examples')` → loads Examples.md
- Tools: `SkillSearch('art tools')` → loads Tools.md
```

--------------------------------

### Install and Mount LifeOS with Hermes

Source: https://docs.ourlifeos.ai/Hermes__HermesSidecar

Install Hermes without Playwright, authenticate with a fresh device-code login, mount LifeOS via the idempotent Mount.ts script, then install and use the launcher. Requires reviewing the install script and pinning to a reviewed SHA. Never import another agent's tokens because OAuth refresh tokens are single-use.

```bash
#!/usr/bin/env bash
set -euo pipefail
# 1. Install Hermes (pinned, no Playwright — LifeOS uses Interceptor)
curl -fsSL https://hermes-agent.nousresearch.com/install.sh -o install.sh
# review it, then:
bash install.sh --non-interactive --skip-setup --skip-browser --commit <reviewed-sha>

# 2. Authenticate (its OWN session — never import a sibling agent's tokens)
hermes auth add openai-codex --type oauth --no-browser

# 3. Mount LifeOS
bun LIFEOS/HERMES/Mount.ts

# 4. Install the launcher (below) and start from it, never from bare `hermes`
```

--------------------------------

### Natural language parsing examples

Source: https://docs.ourlifeos.ai/Pulse__DaSubsystem

Three examples showing how user requests are parsed into scheduled task objects.

```text
User: "remind me at 9am Monday to review PRs"
  -> Parser extracts:
     description: "review PRs"
     schedule: { type: "once", at: "2026-04-07T09:00:00-07:00" }
     action: { type: "notify", message: "Time to review PRs", channel: "voice" }
```

```text
User: "every Friday at 3pm, check newsletter stats"
  -> Parser extracts:
     description: "check newsletter stats"
     schedule: { type: "recurring", cron: "0 15 * * 5" }
     action: { type: "prompt", prompt: "Check UL newsletter stats...", model: "haiku" }
```

```text
User: "in 2 hours, remind me to call the dentist"
  -> Parser extracts:
     description: "call the dentist"
     schedule: { type: "once", at: "<now + 2h>" }
     action: { type: "notify", message: "Reminder: call the dentist", channel: "voice" }
```

--------------------------------

### TruffleHog installation

Source: https://docs.ourlifeos.ai/Tools__Tools

Install TruffleHog via Homebrew. This is a standalone CLI tool that does not require an API key.

```bash
brew install trufflehog
```

--------------------------------

### PULSE.toml example configuration

Source: https://docs.ourlifeos.ai/Pulse__PulseSystem

Example PULSE.toml file showing module configuration sections ([voice], [imessage], [observability], [hooks]) and two [[job]] entries: a script-type job and a claude-type job. Use as a template for defining jobs and enabling modules.

```toml
[voice]
enabled = true
voice_id = "<YOUR_ELEVENLABS_VOICE_ID>"

[imessage]
enabled = false

[observability]
enabled = true

[hooks]
enabled = true

[[job]]
name = "calendar-reminder"
schedule = "*/10 * * * *"
type = "script"
command = "bun run checks/calendar.ts"
output = "voice"
enabled = true

[[job]]
name = "morning-brief"
schedule = "0 7 * * *"
type = "claude"
prompt = "Prepare a morning brief: today's calendar events..."
model = "sonnet"
output = "voice"
enabled = true

```

--------------------------------

### Skill Example 1: Common Use Case

Source: https://docs.ourlifeos.ai/Skills__SkillSystem

Template for a typical skill example showing a user request, the invoked workflow, the skill's action, and the user's result.

```markdown
User: "[Typical user request]"
→ Invokes WorkflowOne workflow
→ [What skill does]
→ [What user gets back]
```

--------------------------------

### Install Work sweep

Source: https://docs.ourlifeos.ai/Services__BackgroundServices

Installs the Work sweep service (com.lifeos.worksweep). This is an opt-in service that runs every hour.

```bash
bun ~/.claude/LIFEOS/TOOLS/InstallWorkSweep.ts
```

--------------------------------

### Install Commitment sweep

Source: https://docs.ourlifeos.ai/Services__BackgroundServices

Installs the Commitment sweep service (com.lifeos.commitmentsweep). This is an opt-in service that runs daily/scheduled.

```bash
bun ~/.claude/LIFEOS/TOOLS/InstallCommitmentSweep.ts
```

--------------------------------

### Verify install with test_guard.py and Mount.ts --check

Source: https://docs.ourlifeos.ai/Hermes__HermesSidecar

Run both checks to verify the install. Trust these results, not the config file. The first script tests deny and allow cases, the second checks identity or config drift.

```shell
python3 LIFEOS/HERMES/plugin/test_guard.py     # deny + allow cases
bun LIFEOS/HERMES/Mount.ts --check             # identity/config drift
```

--------------------------------

### Install Pulse menu-bar app

Source: https://docs.ourlifeos.ai/Services__BackgroundServices

Installs the Pulse menu-bar app service (com.lifeos.pulse-menubar). This is a core service that runs at load.

```bash
bash ~/.claude/LIFEOS/PULSE/MenuBar/install.sh
```

--------------------------------

### Install Inbox sweep

Source: https://docs.ourlifeos.ai/Services__BackgroundServices

Installs the Inbox sweep service (com.lifeos.inboxsweep). This is an opt-in service that runs every 5 minutes.

```bash
bun ~/.claude/LIFEOS/TOOLS/InstallInboxSweep.ts
```

--------------------------------

### Install Atlas asset-graph sync

Source: https://docs.ourlifeos.ai/Services__BackgroundServices

Installs the Atlas asset-graph sync service (com.lifeos.atlas). This is an opt-in service that runs every 15 minutes. The plist is manual at ~/Library/LaunchAgents/com.lifeos.atlas.plist.

```bash
manual plist — ~/Library/LaunchAgents/com.lifeos.atlas.plist (see LIFEOS/DOCUMENTATION/Atlas/AtlasSystem.md)
```

--------------------------------

### Install Pulse dashboard server

Source: https://docs.ourlifeos.ai/Services__BackgroundServices

Installs the Pulse dashboard server service (com.lifeos.pulse). This is a core service that runs at load.

```bash
bash ~/.claude/LIFEOS/PULSE/manage.sh install
```

--------------------------------

### Install Pulse deriver

Source: https://docs.ourlifeos.ai/Services__BackgroundServices

Installs the Pulse deriver service (com.lifeos.deriver). This is a core service that runs daily/scheduled.

```bash
bash ~/.claude/LIFEOS/PULSE/manage-deriver.sh install
```

--------------------------------

### Install a specific service with Services.ts

Source: https://docs.ourlifeos.ai/Services__BackgroundServices

Installs the 'worksweep' service using the Services.ts tool. The --only flag specifies the service, and --yes skips the confirmation prompt. This command is part of the privileged operations that modify launchctl.

```bash
bun ~/.claude/LIFEOS/TOOLS/Services.ts install --only worksweep --yes
```

--------------------------------

### Install Bunker Monitor via bun

Source: https://docs.ourlifeos.ai/Services__BackgroundServices

Command to install the Bunker app monitor service, which runs every 5 minutes to re-run ISA Test Strategy probes and email on green→red/red→green transitions.

```bash
bun ~/.claude/LIFEOS/TOOLS/InstallBunkerMonitor.ts
```

--------------------------------

### Install Conveyor inbox watcher

Source: https://docs.ourlifeos.ai/Services__BackgroundServices

Installs the Conveyor inbox watcher service (com.lifeos.conveyor-watcher). This is an opt-in service that runs at load.

```bash
bun ~/.claude/LIFEOS/TOOLS/InstallConveyorWatcher.ts
```

--------------------------------

### Install Conduit sensory capture

Source: https://docs.ourlifeos.ai/Services__BackgroundServices

Installs the Conduit sensory capture service (com.lifeos.conduit). This is a core service that runs every 2 minutes.

```bash
bun ~/.claude/LIFEOS/PULSE/Conduit/InstallConduit.ts
```

--------------------------------

### Install Conduit insight builder

Source: https://docs.ourlifeos.ai/Services__BackgroundServices

Installs the Conduit insight builder service (com.lifeos.conduit.insight). This is a core service that runs every hour.

```bash
bun ~/.claude/LIFEOS/PULSE/Conduit/InstallConduitInsight.ts
```

--------------------------------

### EXTEND.yaml manifest example

Source: https://docs.ourlifeos.ai/Skills__SkillSystem

Every customization directory requires an EXTEND.yaml manifest. This example shows the required fields: skill name, files to extend, merge strategy, enabled flag, and description.

```yaml
# EXTEND.yaml - Extension manifest
---
skill: SkillName                   # Must match skill name exactly
extends:
  - PREFERENCES.md                 # Files to load
  - OtherConfig.md
merge_strategy: override           # append | override | deep_merge
enabled: true                      # Toggle customizations on/off
description: "What this customization adds"
```

--------------------------------

### state.json example

Source: https://docs.ourlifeos.ai/Pulse__PulseSystem

Example of the Pulse state file located at ~/.claude/Pulse/state/state.json. Written atomically after each job execution. Contains version, startedAt timestamp, and per-job fields: lastRun (Unix ms), lastResult (ok/error), consecutiveFailures counter.

```json
{
  "version": 1,
  "startedAt": 1743451200000,
  "jobs": {
    "email-triage": {
      "lastRun": 1743451500000,
      "lastResult": "ok",
      "consecutiveFailures": 0
    },
    "healthcheck": {
      "lastRun": 1743451500000,
      "lastResult": "error",
      "consecutiveFailures": 2
    }
  }
}
```

--------------------------------

### Agent Prompt Examples for Fast, Standard, and Deep Timing

Source: https://docs.ourlifeos.ai/Delegation__DelegationSystem

Three Agent() calls demonstrating the required ## Scope section for each timing tier: fast (haiku, under 500 words), standard (sonnet, under 1500 words), and deep (opus, no limit). Each example shows the prompt structure and model selection.

```javascript
// FAST — simple check, haiku model, minimal output
Agent({
  prompt: `Check if the auth middleware exports are correct.
## Scope
Timing: FAST — direct answer only.
- Under 500 words
- Answer the question, report the result, done`,
  subagent_type: "Explore",
  model: "haiku"
})

// STANDARD — typical implementation work
Agent({
  prompt: `Senior engineer, TDD. Implement input validation for the login form.
## Scope
Timing: STANDARD — focused implementation.
- Under 1500 words
- Stay on task, deliver the work, verify it works`,
  subagent_type: "general-purpose",
  model: "sonnet"
})

// DEEP — comprehensive analysis
Agent({
  prompt: `Offensive-security operator mindset: map the attack surface, find the seam, prove it, document it. Authorized testing only.
Perform a thorough security review of all auth flows.
## Scope
Timing: DEEP — comprehensive analysis.
- No word limit
- Explore alternatives, consider edge cases
- Thorough verification and documentation`,
  subagent_type: "general-purpose",
  model: "opus"
})
```

--------------------------------

### User extension examples: new .md files at USER/ root

Source: https://docs.ourlifeos.ai/LifeOs__LifeOsSchema

Examples of user extension files that can be added at the USER/ root. Each file must have valid frontmatter; Pulse's fs.watch detects it, the indexer parses it, and the dashboard renders a new tile. No code changes are required in Pulse, Daemon, or Interview.

```markdown
USER/Podcasts.md          # new taste file
USER/Gratitude.md         # new mind file
USER/Travel.md            # new shape file with category: travel (novel category — auto-grouped)
USER/BoardGames.md        # new taste file
```

--------------------------------

### Validation Example

Source: https://docs.ourlifeos.ai/Tools__CliFirstArchitecture

Shows validation flags --dry-run and --explain to preview execution.

```bash
# Validate before executing
evals run --use-case foo --dry-run

# Show what would happen
evals run --use-case foo --explain
```

--------------------------------

### GET /api/wiki

Source: https://docs.ourlifeos.ai/Observability__ObservabilitySystem

Retrieves the system doc index.

```APIDOC
## GET /api/wiki

### Description
System doc index.

### Method
GET

### Endpoint
/api/wiki

### Response
#### Success Response (200)
- System doc index

#### Response Example
{
  "docs": []
}
```

--------------------------------

### Example: Well-Configured Tool (bun Generate.ts)

Source: https://docs.ourlifeos.ai/Skills__SkillSystem

Shows minimal and full invocations of a CLI tool using bun. The full configuration demonstrates resource selection, output configuration, post-processing flags, and output path. Comments in the code explain each flag category.

```bash
# Minimal invocation (sensible defaults)
bun Generate.ts --prompt "..." --output /tmp/image.png

# Full configuration
bun Generate.ts \
  --model nano-banana-pro \    # Resource selection
  --prompt "..." \
  --size 2K \                  # Output configuration
  --aspect-ratio 16:9 \
  --thumbnail \                # Post-processing
  --remove-bg \
  --output /tmp/header.png
```

--------------------------------

### GET /api/security/attack-surface

Source: https://docs.ourlifeos.ai/Observability__ObservabilitySystem

Retrieves deployed-estate scan snapshot. Returns 404 when no scanner is configured on the install.

```APIDOC
## GET /api/security/attack-surface

### Description
Deployed-estate scan snapshot (404 when no scanner is configured on the install).

### Method
GET

### Endpoint
/api/security/attack-surface

### Response
#### Success Response (200)
- Attack surface snapshot

#### Response Example
{
  "snapshot": {}
}

#### Error Response (404)
- No scanner configured
```

--------------------------------

### Install Conduit via launchd

Source: https://docs.ourlifeos.ai/Conduit__ConduitSystem

Commands to install, check status, and uninstall the Conduit launchd agent. The install command polls every 120 seconds.

```bash
bun LIFEOS/PULSE/Conduit/InstallConduit.ts             # install + load (polls every 120s)
bun LIFEOS/PULSE/Conduit/InstallConduit.ts --status    # launchd state
bun LIFEOS/PULSE/Conduit/InstallConduit.ts --uninstall # unload + remove
```

--------------------------------

### Example diary entry

Source: https://docs.ourlifeos.ai/Pulse__DaSubsystem

Example JSON object for a diary entry.

```json
{
  "date": "2026-04-03",
  "interaction_count": 8,
  "topics": ["LifeOS architecture", "newsletter draft", "feed system debugging"],
  "mood": "positive",
  "avg_rating": 8.2,
  "notable_moments": [
    "Designed the DA subsystem together - the principal was excited about heartbeat",
    "Fixed a tricky feed worker bug that had been open for 3 days"
  ],
  "learning": "The principal prefers architecture diagrams in ASCII over Mermaid"
}
```

--------------------------------

### Output Formats Example

Source: https://docs.ourlifeos.ai/Tools__CliFirstArchitecture

Demonstrates human-readable default output, JSON for scripting, and specific fields for parsing.

```bash
# Human-readable by default
evals list use-cases

# JSON for scripting
evals list use-cases --json

# Specific fields for parsing
evals query runs --fields id,score,model
```

--------------------------------

### Cloud Action Execution

Source: https://docs.ourlifeos.ai/Arbol__ArbolSystem

Example curl command to trigger an action in the cloud.

```bash
curl -X POST https://arbol-a-your-action.YOUR-SUBDOMAIN.workers.dev/ \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"content": "Your text here"}'
```

--------------------------------

### Opinions YAML example

Source: https://docs.ourlifeos.ai/Pulse__DaSubsystem

Example YAML file for DA opinions with confidence-weighted beliefs.

```yaml
# DA Opinions -- evolve over time
# Confidence: 0.0 (wild guess) to 1.0 (proven repeatedly)
# Source: observation | inference | stated (principal told me)

opinions:
  - topic: "Architecture documentation format"
    belief: "The principal prefers ASCII diagrams over Mermaid"
    confidence: 0.85
    source: observation
    evidence_count: 7
    first_observed: "2026-01-20"
    last_confirmed: "2026-04-03"

  - topic: "Response length preference"
    belief: "Short for simple tasks, deep when warranted"
    confidence: 0.92
    source: stated
    evidence_count: 12
    first_observed: "2025-12-01"
    last_confirmed: "2026-04-01"

  - topic: "Late night productivity"
    belief: "The principal does their best deep work after 10 PM"
    confidence: 0.70
    source: observation
    evidence_count: 15
    first_observed: "2026-01-05"
    last_confirmed: "2026-03-28"
```

--------------------------------

### Configuration Flags Example

Source: https://docs.ourlifeos.ai/Tools__CliFirstArchitecture

Shows various configuration flags for execution mode, output control, resource selection, and post-processing.

```bash
# Execution mode flags
tool run --fast              # Quick mode (less thorough, faster)
tool run --thorough          # Comprehensive mode (slower, more complete)
tool run --dry-run           # Show what would happen without executing

# Output control flags
tool run --format json       # Machine-readable output
tool run --format markdown   # Human-readable output
tool run --quiet             # Minimal output
tool run --verbose           # Detailed logging

# Resource selection flags
tool run --model haiku       # Use fast/cheap model
tool run --model opus        # Use powerful/expensive model

# Post-processing flags
tool generate --thumbnail    # Generate additional thumbnail version
tool generate --remove-bg    # Remove background after generation
tool process --no-cache      # Bypass cache, force fresh execution
```

--------------------------------

### Skill directory structure with context files

Source: https://docs.ourlifeos.ai/Skills__SkillSystem

Shows the recommended directory layout for a skill using the Dynamic Loading Pattern. Context files (Aesthetic.md, Examples.md, etc.) live directly in the skill root alongside SKILL.md, with Workflows/ and Tools/ subdirectories.

```text
skills/SkillName/ ├── SKILL.md # Minimal routing - loads on invocation ├── Aesthetic.md # Context file - SOP for aesthetic handling ├── Examples.md # Context file - SOP for examples ├── ApiReference.md # Context file - SOP for API usage ├── Tools.md # Context file - SOP for tool usage ├── Workflows/ # Workflow execution files │ ├── Create.md │ └── Update.md └── Tools/ # Actual CLI tools └── Generate.ts
```

--------------------------------

### Arbol CLI Quick Start Commands

Source: https://docs.ourlifeos.ai/Tools__Cli

Commands to run actions with inline JSON, pipe JSON, list actions/pipelines, and show action details. Requires navigating to the Actions directory first.

```bash
cd ~/.claude/LIFEOS/ARBOL/Actions

# Run an action with inline JSON
bun lifeos.ts action A_EXAMPLE_SUMMARIZE --input '{"content": "Your text here"}'

# Pipe JSON into an action
echo '{"content": "Your text"}' | bun lifeos.ts action A_EXAMPLE_SUMMARIZE

# List all available actions
bun lifeos.ts actions

# List all available pipelines
bun lifeos.ts pipelines

# Show action details
bun lifeos.ts info A_EXAMPLE_SUMMARIZE
```

--------------------------------

### Newsletter Parser CLI Migration Example

Source: https://docs.ourlifeos.ai/docs/Tools__CliFirstArchitecture

Shows the before/after of migrating ad-hoc prompting to CLI-First architecture. Step 1 builds the CLI with parser commands; Step 2 updates prompting so the AI executes CLI commands instead of ad-hoc code.

```bash
# Before: Ad-hoc prompting reads/parses/stores content
# After: CLI-First architecture

# Step 1: Build CLI
parser parse --url https://example.com --output content.json
parser store --file content.json --collection newsletters
parser query --collection newsletters --tag ai --limit 10

# Step 2: Update prompting to use CLI
# Instead of ad-hoc code, AI executes CLI commands
```

--------------------------------

### CLI Tool Invocation Examples

Source: https://docs.ourlifeos.ai/docs/Skills__SkillSystem

Shows minimal and full configuration invocations for a well-configured CLI tool. The minimal call relies on sensible defaults, while the full call demonstrates resource, output, and post-processing flags.

```bash
bun Generate.ts --prompt "..." --output /tmp/image.png
```

```bash
bun Generate.ts \
  --model nano-banana-pro \    # Resource selection
  --prompt "..." \
  --size 2K \                  # Output configuration
  --aspect-ratio 16:9 \
  --thumbnail \                # Post-processing
  --remove-bg \
  --output /tmp/header.png
```

--------------------------------

### Install Codex update

Source: https://docs.ourlifeos.ai/Services__BackgroundServices

Installs the Codex update service (com.lifeos.codexupdate). This is an opt-in service that runs daily/scheduled.

```bash
bun ~/.claude/LIFEOS/TOOLS/InstallCodexUpdate.ts
```

--------------------------------

### DA Module Startup Flow Diagram

Source: https://docs.ourlifeos.ai/Pulse__DaSubsystem

ASCII diagram showing the startup sequence for the DA module, including loading modules, checking registry, and registering cron jobs.

```text
Pulse boots
    |
    v
loadModules(config)
    |
    +-- if config.da?.enabled:
    |       |
    |       v
    |   import("./modules/da")
    |       |
    |       v
    |   daModule.startDA(config.da)
    |       |
    |       v
    |   Load DA registry (_registry.yaml)
    |       |
    |       +-- Registry exists?
    |       |       |
    |       |       yes → Load primary DA identity
    |       |       |
    |       |       no → Log warning, start in degraded mode
    |       |            "No DA identity found. Run: bun LIFEOS/TOOLS/DAInterview.ts"
    |       |
    |       v
    |   Register DA cron jobs with Pulse heartbeat loop:
    |       +-- heartbeat check (da-heartbeat)
    |       +-- scheduled task evaluator (da-tasks)
    |       +-- diary writer (da-diary)
    |       +-- growth engine (da-growth)
    |       |
    |       v
    |   DA module running
    |
    v
Continue Pulse startup (HTTP server, other modules, cron loop)
```

--------------------------------

### Example: 3 custom research agents, each a different inline brief

Source: https://docs.ourlifeos.ai/Agents__AgentSystem

Shows how to launch three custom research agents with distinct inline briefs, batched in one message for parallelism.

```javascript
// Example: 3 custom research agents, each a different inline brief
Agent({ subagent_type: "general-purpose", prompt: "You are an enthusiastic, exploratory researcher. …" })
Agent({ subagent_type: "general-purpose", prompt: "You are a skeptical, systematic researcher. …" })
Agent({ subagent_type: "general-purpose", prompt: "You are an analytical, synthesizing researcher. …" })
```

--------------------------------

### Install Conveyor stage engine

Source: https://docs.ourlifeos.ai/Services__BackgroundServices

Installs the Conveyor stage engine service (com.lifeos.conveyor-runner). This is an opt-in service that runs at load.

```bash
bun ~/.claude/LIFEOS/TOOLS/InstallConveyorRunner.ts
```

--------------------------------

### Correct directory structure with context files in skill root

Source: https://docs.ourlifeos.ai/Skills__SkillSystem

Shows the correct directory structure where context files (Aesthetic.md, Examples.md) are placed directly in the skill root, not in a subdirectory.

```text
skills/SkillName/ ├── SKILL.md ├── Aesthetic.md ✅ Context file in skill root └── Examples.md ✅ Context file in skill root
```

--------------------------------

### Action Manifest (action.json)

Source: https://docs.ourlifeos.ai/Arbol__ArbolSystem

Example manifest file defining the action's metadata, schema, and requirements.

```json
{
  "name": "A_LABEL_AND_RATE",
  "description": "Label and rate content using Fabric's label_and_rate pattern.",
  "input": {
    "content": { "type": "string", "required": true },
    "title":   { "type": "string" }
  },
  "output": {
    "one_sentence_summary": { "type": "string" },
    "labels":               { "type": "array" },
    "rating":               { "type": "string" },
    "quality_score":        { "type": "integer" }
  },
  "requires": ["llm", "readFile"]
}
```

--------------------------------

### Progressive Disclosure Example

Source: https://docs.ourlifeos.ai/Tools__CliFirstArchitecture

Shows simple command for common cases and advanced options for complex scenarios.

```bash
# Simple for common cases
evals run --use-case newsletter-summary

# Advanced options available
evals run --use-case newsletter-summary \
  --model claude-3-5-sonnet \
  --prompt v2.0.0 \
  --test-case 001 \
  --verbose \
  --output results.json
```

--------------------------------

### Step 3: Prompting Integration for Evals System

Source: https://docs.ourlifeos.ai/Tools__CliFirstArchitecture

Shows how a user prompt triggers a sequence of CLI commands and the resulting user-visible output.

```text
User: "Run evals for newsletter summary with Claude and GPT-4, then compare them"

AI executes:
1. evals run --use-case newsletter-summary --model claude-3-5-sonnet
2. evals run --use-case newsletter-summary --model gpt-4o
3. evals compare models --use-case newsletter-summary
4. Summarize results in structured format

User sees:
- Run summaries (tests passed, scores)
- Model comparison (which performed better)
- Detailed results if requested
```

--------------------------------

### Deployment procedure: build and restart Pulse

Source: https://docs.ourlifeos.ai/Pulse__PulseSystem

After building the Observatory dashboard, Pulse must be restarted using launchctl stop/start. Do NOT use kill -9 because launchd KeepAlive respawns the process with stale code.

```bash
cd ~/.claude/LIFEOS/Observability && bun run build
launchctl stop com.lifeos.pulse && launchctl start com.lifeos.pulse
```

--------------------------------

### Loading context files via SkillSearch

Source: https://docs.ourlifeos.ai/Skills__SkillSystem

Shows how to load additional context files from workflow files or SKILL.md using SkillSearch calls.

```text
# In workflow files or SKILL.md
SkillSearch('art aesthetic')    # Loads Aesthetic.md from skill root
SkillSearch('art examples')     # Loads Examples.md from skill root
SkillSearch('art tools')        # Loads Tools.md from skill root
```

--------------------------------

### Context file referencing workflows and tools

Source: https://docs.ourlifeos.ai/Skills__SkillSystem

Example of a context file (Aesthetic.md) that references workflows and tools.

```text
# Aesthetic.md (context file)

Use the Essay workflow for blog headers: `Workflows/Essay.md`
Generate images with: `bun Tools/Generate.ts`
```

--------------------------------

### Install Hermes sidecar gateway

Source: https://docs.ourlifeos.ai/Services__BackgroundServices

Installs the Hermes sidecar gateway service (ai.hermes.gateway). This is an opt-in service that runs at load. Health probe: bun LIFEOS/HERMES/Health.ts.

```bash
bun ~/.claude/LIFEOS/HERMES/Mount.ts
```

--------------------------------

### Flow Registry JSON Example

Source: https://docs.ourlifeos.ai/Arbol__ArbolSystem

Example JSON structure for registering a flow in the flow registry. Includes id, name, source, pipeline, destination, and schedule fields.

```json
{
  "flows": [
    {
      "id": "flow-your-source-pipeline",
      "name": "Your Flow Name",
      "source": { "type": "rss", "url": "https://example.com/feed" },
      "pipeline": "P_YOUR_PIPELINE",
      "destination": { "type": "email", "address": "your-email@example.com" },
      "schedule": { "intervalMinutes": 30, "enabled": true }
    }
  ]
}
```

--------------------------------

### Reading a context file directly

Source: https://docs.ourlifeos.ai/Skills__SkillSystem

Shows how to read a specific context file directly using the Read command.

```text
# Read specific context file
Read ~/.claude/skills/Art/Aesthetic.md
```

--------------------------------

### Migration Steps for Primary DA

Source: https://docs.ourlifeos.ai/Pulse__DaSubsystem

Provides the step-by-step migration process for the primary DA, including creating directories, running the migration script, updating configuration, and deprecating old files.

```markdown
Current State:
  LIFEOS/USER/DIGITAL_ASSISTANT/DA_IDENTITY.md          (flat markdown)
  LIFEOS/USER/DA_WRITING_STYLE.md     (separate file)
  LIFEOS_CONFIG.toml → DA section     (voice, personality)

Step 1: Create directory structure
  mkdir -p LIFEOS/USER/DA/your-da

Step 2: Run migration script
  bun LIFEOS/TOOLS/MigrateDAIdentity.ts
  → Reads DA_IDENTITY.md, DA_WRITING_STYLE.md, LIFEOS_CONFIG.toml
  → Writes your-da/DA_IDENTITY.yaml (structured)
  → Writes _registry.yaml (your-da as primary)
  → Generates your-da/DA_IDENTITY.md (for backward compat)
  → Creates empty your-da/growth.jsonl, your-da/opinions.yaml, your-da/diary.jsonl

Step 3: Update LIFEOS_CONFIG.toml
  DA section becomes pointer:
    DA:
      REGISTRY: LIFEOS/USER/DA/_registry.yaml
      PRIMARY: your-da

Step 4: Update ConfigRenderer
  → Reads DA identity from registry path
  → Template variables still resolve

Step 5: Verify
  → CLAUDE.md generates correctly
  → iMessage module uses correct identity
  → Voice uses correct voice_id
  → No visible behavioral change

Step 6: Deprecate old files
  → DA_IDENTITY.md gets header: "# DEPRECATED — See LIFEOS/USER/DA/your-da/DA_IDENTITY.yaml"
  → DA_WRITING_STYLE.md gets header: "# DEPRECATED — Folded into DA_IDENTITY.yaml"
  → Both files kept for 30 days, then deleted

Step 7: Enable DA module
  → Add [da] section to PULSE.toml
  → Restart Pulse
  → Heartbeat starts running
```

--------------------------------

### Agent Session Mapping Example: agent-sessions.json

Source: https://docs.ourlifeos.ai/Hooks__HookSystem

Example content of the agent-sessions.json file that maps session IDs to agent names. This file is used by the agent detection pattern to store and retrieve the agent for each session.

```json
{
  "session-id-abc123": "engineer",
  "session-id-def456": "researcher"
}
```

--------------------------------

### Check learnings directories

Source: https://docs.ourlifeos.ai/Memory__MemorySystem

List contents of the learning system, algorithm, and synthesis directories.

```shell
ls ~/.claude/LIFEOS/MEMORY/LEARNING/SYSTEM/
ls ~/.claude/LIFEOS/MEMORY/LEARNING/ALGORITHM/
ls ~/.claude/LIFEOS/MEMORY/LEARNING/SYNTHESIS/
```

--------------------------------

### Error Handling Example

Source: https://docs.ourlifeos.ai/Tools__CliFirstArchitecture

Shows a clear error message with available use cases and exit codes.

```bash
# Clear error messages
$ evals run --use-case nonexistent
Error: Use case 'nonexistent' not found
Available use cases:
  - newsletter-summary
  - code-review
Run 'evals use-case create' to create a new use case.

# Exit codes
0 = success
1 = user error (wrong args, missing file)
2 = system error (database error, network error)
```

--------------------------------

### GET /voice

Source: https://docs.ourlifeos.ai/Observability__ObservabilitySystem

Retrieves voice status.

```APIDOC
## GET /voice

### Description
Voice status.

### Method
GET

### Endpoint
/voice

### Response
#### Success Response (200)
- Voice status

#### Response Example
{
  "status": "..."
}
```

--------------------------------

### Step 2: CLI Design for Evals System

Source: https://docs.ourlifeos.ai/Tools__CliFirstArchitecture

Provides the CLI commands for managing use cases, test cases, golden outputs, prompts, running evaluations, querying results, and comparing runs.

```bash
# Use case management
evals use-case create --name <name> --description <desc>
evals use-case list
evals use-case show --name <name>
evals use-case delete --name <name>

# Test case management
evals test-case add --use-case <name> --id <id> --input <file>
evals test-case list --use-case <name>
evals test-case show --use-case <name> --id <id>

# Golden output management
evals golden add --use-case <name> --test-id <id> --file <file>
evals golden update --use-case <name> --test-id <id> --file <file>

# Prompt management
evals prompt create --use-case <name> --version <ver> --file <file>
evals prompt list --use-case <name>
evals prompt show --use-case <name> --version <ver>

# Run evaluations
evals run --use-case <name> [--model <model>] [--prompt <ver>]
evals run --use-case <name> --all-models
evals run --use-case <name> --all-prompts

# Query results
evals query runs --use-case <name> [--limit N]
evals query runs --model <model> [--score-min X]
evals query runs --since <date>

# Compare
evals compare runs --run-a <id> --run-b <id>
evals compare models --use-case <name> --prompt <ver>
evals compare prompts --use-case <name> --model <model>
```

--------------------------------

### GET /assistant/opinions

Source: https://docs.ourlifeos.ai/Observability__ObservabilitySystem

Retrieves current DA opinions.

```APIDOC
## GET /assistant/opinions

### Description
Current DA opinions.

### Method
GET

### Endpoint
/assistant/opinions

### Response
#### Success Response (200)
- DA opinions

#### Response Example
{
  "opinions": []
}
```

--------------------------------

### GET /assistant/diary

Source: https://docs.ourlifeos.ai/Observability__ObservabilitySystem

Retrieves recent diary entries.

```APIDOC
## GET /assistant/diary

### Description
Recent diary entries.

### Method
GET

### Endpoint
/assistant/diary

### Response
#### Success Response (200)
- Recent diary entries

#### Response Example
{
  "entries": []
}
```

--------------------------------

### GET /assistant/avatar

Source: https://docs.ourlifeos.ai/Observability__ObservabilitySystem

Retrieves the DA avatar image.

```APIDOC
## GET /assistant/avatar

### Description
DA avatar image.

### Method
GET

### Endpoint
/assistant/avatar

### Response
#### Success Response (200)
- Avatar image data

#### Response Example
{
  "image": "..."
}
```