Team Kb
Answer questions about the team's knowledge base using DocMind. Use this skill whenever someone asks about team documentation, project specs, architecture decisions, technical requirements, API specs, or any content stored in the team knowledge base. Trigger on questions like "what does the spec say about X?", "find information about Y in our docs", "how does Z work according to our documentation…
Install
$ agentstack add skill-tano73-agent-skills-team-kb ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Links to your public security report.[](https://agentstack.voostack.com/security/report/skill-tano73-agent-skills-team-kb) Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Team Knowledge Base
Answer questions by searching the team knowledge base. Never invent information — every claim must come from a retrieved document.
Step 1 — Check the local LLM wiki (if available)
If the llm-wiki-manager skill is available and a wiki/ directory exists in the workspace, query it first. It may already contain a synthesized answer.
Step 2 — Search DocMind
Call searchFlavorChunks with these parameters:
| Parameter | Value | |-----------|-------| | project | Project name from the user. If unknown, call listProjects first and pick the most relevant one. | | query | Rephrase the user's question in 3–8 keywords; expand abbreviations; add synonyms. | | mode | "hybrid" (default). Use "fulltext" for exact terms or IDs. | | adjacentChunks | 1 or 2 when a chunk looks cut off or needs surrounding context. | | limit | 5–8 chunks; raise to 10 for broad topics. |
If results are poor or empty, retry in order:
- Rephrase the query with different keywords.
- Switch to
mode: "semantic". - Try a related project with
listProjects.
Step 3 — Build and format the answer
- Write only what the retrieved chunks support. Do not add outside knowledge.
- Cite every key point inline:
(Source: , line N)orlines N–M. - If chunks from multiple documents support the same point, cite all of them.
- For factual questions: direct answer → supporting quote → citation.
- For broad questions: use headings or bullets, each backed by a citation.
- End every response with a Sources section listing all document names used.
When information is insufficient
If no search attempt returns enough content, do not guess. Ask the user:
> "Non ho trovato informazioni sufficienti nella knowledge base. Vuoi che estenda la ricerca a: > - 🌐 Web (ricerca pubblica online)? > - 🧠 Base di conoscenza interna (conoscenza generale del modello)? > > In entrambi i casi indicherò sempre le fonti utilizzate."
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: Tano73
- Source: Tano73/agent-skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.