solaceproducts/solace-agent-mesh/support-playbook
- community
Checking that the SKILL.md file is still in the repository.
The support team's rules for the first reply to a customer, plus its knowledge-base articles. Load it before drafting or reviewing any reply to a customer issue.
Install
npx vskill@latest install solaceproducts/solace-agent-mesh/support-playbookSKILL.md
Support playbook
Follow these rules for every first reply to a customer.
Before you write
- Search the knowledge base for the customer's problem with
grep_skill_resources (try two or three key words from the issue, such as invoice, CSV, charge, 503, password), then read any article that matches with read_skill_resource. Articles are under references/kb/.
- If an article matches, the reply must use it: give its steps or its policy
in your own words. If no article matches, do not invent product steps, menus or policies.
What every reply contains
- One sentence that restates the problem in the customer's own terms.
- When they will hear from us next, as one sentence such as "We will get back
to you by <deadline>." The deadline is our commitment to the customer, never a deadline for the customer. Copy it exactly as you were given it; if you were given none, say the team will follow up shortly. Never work out a deadline yourself, and never promise a fix time or a root cause.
- The answer, if a knowledge-base article covers it.
- Every item from the triage's missing information, as a numbered list. Skip
this list if there is nothing missing.
- The sign-off "The Support Team". No subject line.
Rules by category
- Outage: apologise once, point to the status page in
references/kb/service-status.md, and say updates follow there.
- Billing: never ask for a full card number, password or bank details.
Follow references/kb/duplicate-charges.md for duplicate charges.
- Account access: never ask for a password. Follow
references/kb/password-reset.md.
- How-to: if an article answers the question, the reply is the answer.
Keep it to the steps the article gives.
Tone
Plain, warm and short: at most 150 words before the numbered list. No exclamation marks, no "we apologise for any inconvenience", no internal team names, priorities or severity labels.
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingexamples/support-triage/agent-mesh-config/skills/support-playbook/SKILL.md
Popularity
- Repository stars
- 4,913
- Repository forks
- 274
- 7-day trend score
- 0
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/solaceproducts/solace-agent-mesh/support-playbook)