More screenshots
Works with the AI Agents you already use
Enterprise Knowledge Gap Detection Agent
Detects missing, conflicting, and outdated enterprise knowledge using support tickets, search logs, and document audits.
$7.99
Enterprise Knowledge Gap Detection Agent
Example session with this skill installed
Analyze these support ticket logs and our internal FAQ. Identify where agents are giving answers not found in the documentation and prioritize the gaps by ticket volume.
- Read your context and instructions
- Compiled the enterprise knowledge gap
Knowledge Gap Report
- Gap: Recovering MFA after device loss.
- Evidence: 42 tickets this week; 0 matches in FAQ.
- Current State: Agents use unofficial chat notes.
- Priority: P1 (High Demand).
- Action: Create 'MFA Recovery Guide' using the attached agent notes as a draft.
Connects securely to your tools. The creator never sees your data.
About this skill
The problem
Large organizations often have vast content repositories yet struggle to provide consistent answers. Employees waste time searching for missing or fragmented info, while customers receive conflicting instructions from different support channels.
What it does
- Analyzes support tickets, search logs, and employee questions to extract real-world information demand.
- Maps existing documents against user questions to identify missing, outdated, or contradictory answers.
- Detects knowledge fragmentation where answers are split across multiple repositories or silos.
- Performs findability audits to reveal why users can't locate existing content using natural language.
- Generates prioritized remediation backlogs with risk scores and ownership assignments.
Why this beats prompting it yourself
Generic prompts often miss the distinction between missing content and poor findability or hidden permissions. This tool uses a structured evidence model to separate facts from inference, ensuring audits are based on actual query logs and ticket patterns rather than just document counts.
Use cases
- Auditing internal wikis and handbooks for stale or conflicting policies after a merger.
- Identifying the cause of zero-result searches and high-volume support escalations.
- Evaluating content readiness before deploying RAG-based AI assistants to employees.
- Mapping undocumented exceptions that support teams handle but never record officially.
Known limitations
Requires structured exports from support systems or search logs to provide high-confidence demand scoring. It cannot independently authorize policy or assign official ownership without human review.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse All Claude Skills.
Requires source data like ticket exports or search logs. Cannot independently approve policy or verify search outcomes without authorized expert review.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 3 months ago
Permissions
File Scopes
This skill uses file access to read user-provided internal documents, policies, procedures, standard operating procedures, training content, knowledge-base articles, support tickets, help-desk cases, employee questions, customer questions, search logs, zero-result reports, internal chat exports, incident reports, quality findings, onboarding materials, analytics exports, content inventories, ownership records, metadata, permission information, governance standards, client feedback, and QA notes. It uses write access to create source inventories, authority maps, topic clusters, demand analyses, search-gap reports, ticket-derived gap reports, employee-question analyses, contradiction reports, staleness reports, duplication reports, fragmentation reports, findability audits, permission-gap reports, ownership-gap reports, AI-readiness assessments, remediation backlogs, content briefs, governance recommendations, executive summaries, operational reports, acceptance criteria, QA checklists, risk assessments, and SKILL.md files. Browser access is useful when the host environment permits authorized repository access, public product-documentation review, public regulatory-source review, public standards research, approved terminology research, or current platform-documentation validation. The skill should clearly distinguish: Observed: Directly supported by documents, tickets, employee questions, search logs, analytics, repository metadata, system records, or approved governance information. Inferred: A likely knowledge gap, cause, relationship, audience need, operational impact, or remediation opportunity suggested by available evidence. Recommended: A proposed content update, new article, merge, archive, metadata change, search improvement, permission review, ownership assignment, governance control, training update, or AI-index action. Requires Verification: Any authority decision, policy interpretation, legal requirement, permission change, owner assignment, risk classification, deletion action, archival decision, causal claim, or final approved answer requiring confirmation. Conceptual: A proposed enterprise knowledge-management workflow, taxonomy, governance system, or AI-readiness architecture that has not yet been approved or implemented. The agent must not silently reconcile conflicting policies, choose an authority without evidence, delete content, change permissions, archive records, or assign ownership without approval. A recently modified document is not automatically current. A closed support ticket is not automatically evidence of successful resolution. A search click is not automatically evidence that the user found the answer. A frequently used document is not automatically accurate. A low-use document is not automatically unnecessary. Public examples should use fictional, anonymized, masked, aggregated, or approved organizational data. Sensitive, confidential, regulated, privileged, security-restricted, employee, and customer information should be handled according to approved access and retention rules.