# Managing Saltstack

> Use when working with Saltstack — saltStack configuration management and remote execution. Covers state management, grain data, pillar data, job management, targeting, event system, and orchestration. Use when managing Salt infrastructure, executing remote commands, debugging state failures, or inspecting minion data.

- Skill: `cloudthinker-ai/managing-saltstack` (Agent Skill)
- Install (CLI): `npx skillmds@latest add cloudthinker-ai/managing-saltstack`
- Raw SKILL.md: https://api.skillmd.com/api/skills/cloudthinker-ai/managing-saltstack/raw
- Safety review: pending (external: skill-scanner PASS, skillspector CAUTION)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: cloudthinker-ai (https://skillmd.com/u/cloudthinker-ai)
- Updated: 2026-09-08
- Page: https://skillmd.com/skills/cloudthinker-ai/managing-saltstack

---


# SaltStack Management Skill

Manage and inspect SaltStack states, grains, pillars, jobs, and minion connectivity.

## MANDATORY: Discovery-First Pattern

**Always check minion connectivity and key status before executing states or commands.**

### Phase 1: Discovery

```bash
#!/bin/bash

echo "=== Salt Version ==="
salt --version 2>/dev/null

echo ""
echo "=== Minion Key Status ==="
salt-key --list all 2>/dev/null | head -20

echo ""
echo "=== Connected Minions ==="
salt '*' test.ping --output=json 2>/dev/null | jq 'to_entries | length' | xargs -I{} echo "{} minions responding"

echo ""
echo "=== Salt Master Status ==="
salt-run manage.status 2>/dev/null | head -15
```

## Core Helper Functions

```bash
#!/bin/bash

# Salt command wrapper with timeout
salt_cmd() {
    salt "$@" --timeout=30 --output=json 2>/dev/null
}

# Salt API call (if salt-api is running)
salt_api() {
    local endpoint="$1"
    local data="${2:-}"
    curl -s -H "X-Auth-Token: $SALT_TOKEN" \
        -H "Content-Type: application/json" \
        "http://localhost:8000/$endpoint" \
        ${data:+-d "$data"}
}

# Target helper
salt_target() {
    local target="$1"
    shift
    salt "$target" "$@" --timeout=30 --output=json 2>/dev/null
}
```

## Output Rules
- **TOKEN EFFICIENCY**: Target <=50 lines per output
- Use `--output=json` with jq for structured output
- Use compound targeting to narrow scope
- Never run commands against '*' in production without `--batch-size`

## Common Operations

### State Management

```bash
#!/bin/bash
TARGET="${1:?Target required (minion or glob)}"
SLS="${2:-}"

if [ -n "$SLS" ]; then
    echo "=== Dry Run: State $SLS on $TARGET ==="
    salt "$TARGET" state.apply "$SLS" test=True --output=json 2>/dev/null | jq '
        to_entries[] | {
            minion: .key,
            states: [.value | to_entries[] | {
                id: .key,
                result: .value.result,
                changes: (.value.changes | length > 0),
                comment: .value.comment
            }]
        }
    ' | head -50
else
    echo "=== Highstate Dry Run on $TARGET ==="
    salt "$TARGET" state.highstate test=True --output=json 2>/dev/null | jq '
        to_entries[] | {
            minion: .key,
            summary: {
                total: (.value | length),
                changed: ([.value | to_entries[] | select(.value.changes | length > 0)] | length),
                failed: ([.value | to_entries[] | select(.value.result == false)] | length)
            }
        }
    ' | head -30
fi
```

### Grain and Pillar Inspection

```bash
#!/bin/bash
TARGET="${1:?Target required}"

echo "=== Key Grains ==="
salt "$TARGET" grains.items --output=json 2>/dev/null | jq '
    to_entries[] | {
        minion: .key,
        os: .value.os,
        osrelease: .value.osrelease,
        kernel: .value.kernel,
        cpus: .value.num_cpus,
        mem_total: .value.mem_total,
        ip4: .value.ipv4
    }
' | head -30

echo ""
echo "=== Pillar Data ==="
salt "$TARGET" pillar.items --output=json 2>/dev/null | jq '
    to_entries[] | {minion: .key, pillar_keys: (.value | keys)}
' | head -20
```

### Job Management

```bash
#!/bin/bash
echo "=== Active Jobs ==="
salt-run jobs.active --output=json 2>/dev/null | jq '
    to_entries[] | {
        jid: .key,
        function: .value.Function,
        target: .value.Target,
        user: .value.User,
        started: .value.StartTime
    }
'

echo ""
echo "=== Recent Jobs ==="
salt-run jobs.list_jobs --output=json 2>/dev/null | jq '
    to_entries | sort_by(.key) | reverse | .[:10][] | {
        jid: .key,
        function: .value.Function,
        target: .value.Target,
        user: .value.User
    }
'

echo ""
echo "=== Job Result ==="
JID="${1:-}"
if [ -n "$JID" ]; then
    salt-run jobs.lookup_jid "$JID" --output=json 2>/dev/null | jq '.' | head -30
fi
```

### Remote Execution

```bash
#!/bin/bash
TARGET="${1:?Target required}"
CMD="${2:?Command required}"

echo "=== Executing on $TARGET ==="
salt "$TARGET" cmd.run "$CMD" --timeout=30 --output=json 2>/dev/null | jq '
    to_entries[] | "\(.key): \(.value)"
' | head -30
```

### Event and Reactor Monitoring

```bash
#!/bin/bash
echo "=== Recent Events ==="
salt-run state.event tagmatch='salt/*' count=10 2>/dev/null | head -20

echo ""
echo "=== Reactor Configuration ==="
cat /etc/salt/master.d/reactor.conf 2>/dev/null || \
grep -A 10 'reactor:' /etc/salt/master 2>/dev/null | head -15
```

## Safety Rules

- **NEVER run highstate on '*' in production** -- use `--batch-size` or specific targeting
- **Always use `test=True` first** before applying states
- **Pillar data may contain secrets** -- never dump all pillar data without filtering
- **Key acceptance is a security operation** -- verify minion identity before accepting keys
- **Use compound targeting** for precision -- avoid broad globs in production

## Output Format

Present results as a structured report:
```
Managing Saltstack Report
═════════════════════════
Resources discovered: [count]

Resource       Status    Key Metric    Issues
──────────────────────────────────────────────
[name]         [ok/warn] [value]       [findings]

Summary: [total] resources | [ok] healthy | [warn] warnings | [crit] critical
Action Items: [list of prioritized findings]
```

Target ≤50 lines of output. Use tables for multi-resource comparisons.

## Anti-Hallucination Rules

1. **NEVER assume resource names** — always discover via CLI/API in Phase 1 before referencing in Phase 2.
2. **NEVER fabricate metric names or dimensions** — verify against the service documentation or `--help` output.
3. **NEVER mix CLI commands between service versions** — confirm which version/API you are targeting.
4. **ALWAYS use the discovery → verify → analyze chain** — every resource referenced must have been discovered first.
5. **ALWAYS handle empty results gracefully** — an empty response is valid data, not an error to retry.

## Counter-Rationalizations

| Shortcut | Counter | Why |
|----------|---------|-----|
| "I'll skip discovery and check known resources" | Always run Phase 1 discovery first | Resource names change, new resources appear — assumed names cause errors |
| "The user only asked for a quick check" | Follow the full discovery → analysis flow | Quick checks miss critical issues; structured analysis catches silent failures |
| "Default configuration is probably fine" | Audit configuration explicitly | Defaults often leave logging, security, and optimization features disabled |
| "Metrics aren't needed for this" | Always check relevant metrics when available | API/CLI responses show current state; metrics reveal trends and intermittent issues |
| "I don't have access to that" | Try the command and report the actual error | Assumed permission failures prevent useful investigation; actual errors are informative |

## Common Pitfalls

- **Minion not responding**: Check minion service, firewall (ports 4505/4506), and key acceptance
- **Pillar rendering errors**: Jinja errors in pillar files cause empty/partial pillar data -- check `pillar.items` for errors
- **State ordering**: States execute in file order by default -- use `require`/`watch` for explicit ordering
- **Grain targeting drift**: Grain values can change -- don't rely on volatile grains for targeting
- **Job timeout**: Long-running states may exceed default timeout -- increase with `--timeout`
- **Returner failures**: If returner (MySQL, Redis) is down, job results may be lost
- **File server caching**: Salt master caches files -- use `salt-run fileserver.update` after changes
- **Reactor storms**: Poorly scoped reactors can trigger cascading events -- test reactor configs carefully