gutt memory onboarding

VerifiedSafe

First-use onboarding for gutt memory integration. Verifies MCP connection, demonstrates memory search and capture, and explains the HUD.

Sby Skills Guide Bot
ProductivityBeginner
8/5/2026
Claude Code
#onboarding#memory#mcp#knowledge-graph#setup

Recommended for

Automated summary

Walks a user through first-time setup of the gutt memory integration: it checks MCP configuration and connectivity, demonstrates a memory search and a first capture, then explains the optional HUD and the active hooks.

Strengths

  • Detects the MCP configuration in project settings and gives a troubleshooting table covering unconfigured, unreachable, and auth-failure cases.
  • Demonstrates memory hands-on: one graph search, then a first capture recording that the plugin was activated for the project.
  • Explicitly treats an empty memory graph as expected on a fresh setup, avoiding false alarms.
  • For the HUD, has the user run /gutt-pro:statusline themselves instead of silently writing to their settings file.

Limitations

  • The whole flow depends on a reachable gutt MCP server; without one it halts at the connectivity check.
  • Scope is limited to first-run onboarding — actual configuration is deferred to /gutt-pro:setup rather than explained.
  • The active-hooks section is truncated in the supplied file, so the full hook list cannot be verified.
When to use it

Use it when someone has just installed gutt in a project and wants to confirm the memory connection works end to end.

When not to use it

Don't use it to debug an already-running MCP server or to set up groups and API keys, which belong to the dedicated commands.

Security analysis

Safe
Quality score85/100

The skill only reads local settings files with a harmless grep check and calls MCP functions for memory search/capture. No destructive commands, exfiltration, or obfuscation are present. It guides the user through configuration but does not execute arbitrary network payloads or disable safety.

No concerns found

How this analysis is made

Examples

First-time setup
I just installed gutt in this project — walk me through the setup and check that the MCP server is actually connected.
How does this work?
First time using this plugin. How does the gutt memory thing work, and what's already in my memory graph?
Activate memory
Onboard me to gutt memory and capture a first entry so I can see it working.

name: onboard description: "First-use onboarding for gutt memory integration. Verifies MCP connection, demonstrates memory search and capture, explains HUD and active hooks. Triggers on: onboard, getting started, first time, setup guide, how does this work."

Onboard Skill

Announce: "Starting gutt memory onboarding..."

Walk the user through first-use setup and verification of the gutt memory integration.

Flow

Step 1: Detect MCP

Check if the gutt MCP server is configured in the project's settings.

# Check for gutt MCP configuration in settings
cat .claude/settings.json 2>/dev/null | grep -q "gutt"
  • If configured: Announce "gutt MCP server found in settings." and proceed to Step 2.
  • If NOT configured: Ask the user for their MCP server URL and guide them through setup:
    1. Ask: "What is your gutt MCP server URL? (e.g., https://your-instance.gutt.io/mcp)"
    2. Explain they can also run /gutt-pro:setup to configure it.
    3. Once URL is provided, help add it to .claude/settings.json.

Step 2: Verify Connectivity

Run a test query to confirm the MCP connection is live:

search_memory_nodes(query="test connectivity", max_nodes=1)

| Result | Action | | ---------------------- | ------------------------------------------------------------------ | | Success (any response) | "MCP connection verified." Proceed to Step 3. | | Not configured error | "MCP server not configured. Run /gutt-pro:setup first." Stop. | | Unreachable / timeout | "MCP server unreachable. Check your network and server URL." Stop. | | Auth failure | "Authentication failed. Check your API key or token." Stop. |

Step 3: First Search

Demonstrate memory retrieval by searching the knowledge graph:

search_memory_nodes(query="recent decisions or lessons", max_nodes=5)

Show results formatted as:

## Your Memory Graph

Found [N] entries:

| Entity | Type   | Summary         |
| ------ | ------ | --------------- |
| [name] | [type] | [brief summary] |

Explain: "The gutt memory graph stores organizational knowledge -- lessons learned, decisions made, patterns discovered, and relationships between them. Every search, capture, and agent delegation flows through this graph."

If no results are returned: "Your memory graph is empty -- that is expected for a new setup. It will grow as you work."

Step 4: First Capture

Demonstrate memory capture by storing an onboarding record:

add_memory(
  name="Onboarding: plugin activated",
  episode_body="gutt-pro onboarding completed for this project. Plugin active with memory integration.",
  source="text",
  source_description="onboard skill - first capture"
)

Confirm: "First memory captured. This entry marks the activation of the plugin for this project."

Step 5: HUD Explainer

Offer the status bar HUD. It is off until the user asks for it — a plugin cannot install a status line, so /gutt-pro:statusline is the only way it appears, and it writes the user's own settings file only when they run it.

An optional HUD can show gutt's state in your status bar:

  [gutt 🟢 on <group>]                   configured, recall live
  [gutt ⚪ on !]                         no gutt MCP server — run /gutt-pro:setup
  [gutt 🟢 off]                          recall disabled — /gutt-pro:on restores it
  [gutt 🟢 zzz→14:30 <group>]            snoozed until then

Turn it on with /gutt-pro:statusline, off again with /gutt-pro:statusline off.

If they accept, tell them to run the command themselves rather than running it for them — the point of the command is that installing it is their decision.

Step 6: Active Hooks

List all hooks that are active and what they do:

## Active Hooks

| Hook                 | Event                | What It Does                                           |
| -------------------- | -------------------- | ------------------------------------------------------ |
| session-start        | SessionStart         | Opens the session record and runs the state TTL sweep  |
| session-connectivity | SessionStart (async) | Probes MCP configuration for the HUD                   |
| session-end          | SessionEnd           | Finalizes the session record and clears session snooze |
| user-prompt-submit   | UserPromptSubmit     | Injects relevant memory context before each prompt     |
| post-memory-search   | PostToolUse (gutt)   | Records that memory was searched, for the recency gate |
| _(prompt hook)_      | Stop                 | Judges whether the turn is worth capturing to memory   |

The status bar HUD is not a hook and is not on by default — a plugin cannot install one. Mention that /gutt-pro:statusline turns it on, and that it shows connection state, whether recall is on, off or snoozed, and the group being written to.

Step 7: Next Steps

## Next Steps

You are ready to use gutt memory integration. Here is what to try:

1. **Search before tasks** -- Use `/memory-search` or just start working;
   the hooks automatically search memory for relevant context.

2. **Capture corrections and decisions** -- Use `/memory-capture` when you
   learn something worth remembering, or say "remember that..." naturally.

3. **Available agents** -- Check `agents/` for specialized agents like
   `plugin-dev`, `plugin-qa`, `hook-expert`, and more.

4. **Skills** -- Run `/skills-discovery` to analyze what additional skills
   could help your workflow.

--check Mode

When invoked with --check, skip the tutorial and just verify:

  1. MCP connectivity -- Run test query, report pass/fail.
  2. Hook registration -- Confirm hooks.json is loadable and all hook scripts exist on disk.
  3. Statusline -- Look in ~/.claude/settings.json, the only place a status line can live. Report it as installed, someone else's, or absent. Absent is not a fault; it is the default until the user runs /gutt-pro:statusline.

Output a compact status table:

## gutt Plugin Health Check

| Component         | Status                                     |
| ----------------- | ------------------------------------------ |
| MCP connectivity  | OK / FAIL: [reason]                        |
| Hook registration | OK ([N] hooks) / FAIL: [reason]            |
| Statusline        | installed / not installed / someone else's |

From the file to a first result

The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.

View the source and supporting files

This download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.

1. Download the file

Download SKILL.md

From the root of a test project, save the file as .skill-downloads/gutt-memory-onboarding-msfn1a6k/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.

Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.

mkdir -p '.skill-downloads/gutt-memory-onboarding-msfn1a6k' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/gutt-memory-onboarding-msfn1a6k/download' --output '.skill-downloads/gutt-memory-onboarding-msfn1a6k/SKILL.md.part' && mv '.skill-downloads/gutt-memory-onboarding-msfn1a6k/SKILL.md.part' '.skill-downloads/gutt-memory-onboarding-msfn1a6k/SKILL.md'

2. Check prerequisites

Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.

No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.

The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.

3. Load the skill

After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.

Claude Code
test -f '.skill-downloads/gutt-memory-onboarding-msfn1a6k/SKILL.md' && test -s '.skill-downloads/gutt-memory-onboarding-msfn1a6k/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/onboard' && cp -R '.skill-downloads/gutt-memory-onboarding-msfn1a6k' '.claude/skills/onboard'

OpenSkills

OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.

Official OpenSkills installation
openskills install './.skill-downloads/gutt-memory-onboarding-msfn1a6k'

openskills read 'gutt-memory-onboarding-msfn1a6k'

Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.

For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.

openskills sync

Manual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.

4. Try a first task

Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:

Read .skill-downloads/gutt-memory-onboarding-msfn1a6k/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.

Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.

Related skills