Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    expertise directory inference

    1

    Identifying who knows what in a large organization is often manual, outdated, or reliant on guesswork.

    $1.99

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    expertise directory inference

    expertise directory inference

    Example session with this skill installed

    Build an expertise directory from the provided employee profiles and document metadata. Flag any topics where we only have one expert.

    • Read your context and instructions
    • Compiled the expertise directory inference

    The analysis is complete. I have processed 12 employee profiles and 45 authored assets to generate this draft directory. Two profiles were excluded due to missing consent flags.

    Expertise Directory Draft — Engineering Dept, 2024-05-20

    1. Summary

    • Profiles Analyzed: 10 (2 excluded)
    • Total Assets: 45
    • Unique Topics: 8
    • Concentration Risks: 2

    2. Directory entries (validation required)

    NameDepartmentPrimary TopicExpertise ScoreStatus
    Sarah ChenPlatformKubernetes0.88DRAFT
    Marcus V.SecurityOAuth20.92DRAFT
    Elena RossiDevOpsCI/CD0.75DRAFT

    3. Topic Coverage

    TopicExpert CountAverage Score
    Kubernetes30.72
    OAuth210.92
    Terraform40.65

    4. Concentration Risks

    • Single Expert: OAuth2 (Expert: Marcus V.). High risk if the individual leaves.
    • Dominant Expert: CI/CD (Elena Rossi holds 80% of authored assets).

    5. Excluded profiles

    • Count: 2 (Consent not provided).

    6. Prioritized actions

    ActionPriorityReason
    Validate Marcus V. for OAuth2HighSingle expert risk
    Cross-train on CI/CDMediumKnowledge concentration

    7. Warnings and items for human review

    Management decision required for publication. Verify that authored asset metadata correctly reflects current roles.

    Next steps

    1. Review the Concentration Risks in section 4 to plan knowledge transfer.
    2. Confirm the 2 excluded profiles have been contacted regarding consent.
    3. Validate the expertise scores in section 2 before internal distribution.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Identifying who knows what in a large organization is often manual, outdated, or reliant on guesswork. This skill automates the creation of an expertise directory by analyzing employee profiles and their actual authored assets to infer subject matter expertise.

    What it does

    • Expertise Inference calculates scores based on authored content and metadata with recency weighting.
    • Topic Coverage Mapping generates a directory showing which employees are authorities on specific subjects.
    • Concentration Risk Analysis identifies single points of failure where expertise is limited to one or two people.
    • Consent Enforcement automatically filters out employee profiles that have not provided explicit consent for profiling.
    • Action Prioritization suggests specific validation steps and knowledge transfer tasks for management review.

    How it works

    1. Data ingestion reads employee profiles and metadata from authored assets like documents or reports.
    2. Request generation builds a structured JSON input, strictly including only those with verified consent flags.
    3. Deterministic analysis runs a core script to compute expertise scores and identify knowledge risks.
    4. Draft interpretation processes the result envelope to produce a validated, formatted report for human review.

    Frameworks & tools

    The skill utilizes a Python-based calculation engine and operates with JSON schemas for data validation. It integrates with Markdown for report generation.

    Why this beats prompting it yourself

    Generic prompts often hallucinate expertise based on job titles or resumes. This skill uses a deterministic script and recency-weighted formulas to ensure scores are based on actual output, while strictly enforcing data privacy and consent rules that a standard prompt might bypass.

    Use cases

    • Build a draft expertise directory for a department or entire organization.
    • Find subject matter experts for a specific project or technical query.
    • Identify "single-expert" topics that pose a knowledge concentration risk.
    • Analyze topic coverage across teams to assist in knowledge management planning.

    Known limitations

    Does not infer skills from job titles or resumes. Requires structured input with explicit consent flags for each profile. Results are drafts that require human validation before publication.

    Requires structured profile data with consent flags. Does not infer skills from job titles. Analysis is limited to authored content metadata.

    How to use

    1. 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. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    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 1 day ago

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Environment Variables
    Listed1 day ago

    What's inside

    Frequently Asked Questions

    Unlike generic AI prompts that guess expertise based on resumes, this skill uses a deterministic, recency-weighted calculation engine to analyze actual authored content, ensuring results reflect current, active knowledge rather than outdated job titles.

    The skill is built to be LLM-agnostic, requiring only that your AI agent can process JSON data and execute the included Python-based analysis logic. It is compatible with major agentic frameworks that support structured data ingestion and tool calling.

    Privacy is baked into the logic; the skill requires a mandatory 'consent flag' for every profile and will automatically exclude any individual from the analysis who hasn't explicitly opted in, ensuring your internal directory remains compliant.

    You will receive the core Python calculation engine, the JSON schema for data validation, and the reporting templates, providing everything needed to transform raw asset metadata into a formatted expertise risk report.

    No, this skill focuses strictly on 'proof of work'—it analyzes authored documents and assets to verify expertise, intentionally ignoring resumes and job titles to avoid reporting bias or aspirational skill listing.

    The output includes a 'Concentration Risk' report, which highlights specific topics where only one or two people are authorities, allowing management to prioritize knowledge transfer before those individuals leave or change roles.