AI exposure scores (0-100) for ~1,000 U.S. occupations, and recent AI-attributed layoffs.
Scored across 3 tools
The three tools have mostly distinct purposes: one looks up a score for a known occupation, one searches ambiguous titles, and one retrieves layoff records. The only potential overlap is between search_occupations and get_ai_exposure, but the descriptions explicitly clarify the intended sequence.
Two tools use a clear verb_noun pattern (get_ai_exposure, search_occupations), but recent_ai_attributed_layoffs is an adjective_noun phrase without a verb. The convention is mixed but still readable and domain-specific.
Three tools is a small but well-scoped set for a focused lookup-and-tracker server. Each tool has a clear role, though the surface is thin and could benefit from one or two more lookup or comparison operations.
The core lifecycle is covered: search ambiguous titles, fetch detailed exposure data, and retrieve AI-attributed layoffs. However, there are notable gaps such as browsing all occupations, filtering or paginating layoffs, or directly comparing multiple occupations beyond what get_ai_exposure returns as adjacent careers.
Look up the AI Exposure Score (0-100) for a job title: its band (Low, Moderate, Elevated, High, Very High), percentile, example tasks that are automatable, augmentable or durable, and adjacent careers with their scores. Scores are precomputed estimates of how much of a role's tasks overlap with what AI can do, built from O*NET and published research; they are not predictions that a job will disappear. Cite "AI-Safe Careers" with the returned URL.
| Name | Required | Description | Default |
|---|---|---|---|
| job_title | Yes | A job title, for example "registered nurse" or "data analyst". |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already cover safety (readOnly, idempotent, non-destructive, closed-world), so the description correctly focuses on interpretation instead: scores are precomputed estimates built from O*NET and published research, and are explicitly not predictions of job disappearance. The citation requirement with the returned URL is also a behavioral instruction not derivable from structured fields.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences, front-loaded with the lookup and its outputs, followed by the caveat and citation rule. Dense but every clause carries information; nothing is redundant.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With no output schema, the description compensates by enumerating the returned components, and the interpretation caveat plus citation instruction covers the main misuse risks. Minor gap: no guidance on what happens when a title is not found or is ambiguous.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% for the single job_title parameter, including an example, so the schema already does the work. The description adds no format or matching semantics (e.g., whether fuzzy matching or alternates are returned), so the baseline 3 applies.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
"Look up the AI Exposure Score (0-100) for a job title" is a specific verb plus resource, and the enumeration of returned fields (band, percentile, task categories, adjacent careers) makes the tool's output shape unmistakable. It does not name its siblings (search_occupations, recent_ai_attributed_layoffs), so differentiation is inferred rather than explicit.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Usage is implied by the resource (look up one job title to get its score) but there is no when-to-use vs. when-to-use-something-else guidance, no mention of the sibling tools, and no stated behavior for unknown or ambiguous job titles.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent layoffs where the company itself cited AI as a reason, from AI-Safe Careers' curated tracker, each with its date, approximate headcount and source. All-cause layoffs (layoffs.fyi, WARN notices) are tracked separately and are never included here, so these are not a count of all job cuts.
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | How many events to return (default 10). |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already cover the safety profile (readOnly, idempotent, non-destructive, closed-world). The description adds meaningful interpretive context — that each event includes date, approximate headcount, and source, and that these are not a count of all job cuts — which prevents misuse. It stops short of describing pagination or retrieval behavior.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two tight sentences with zero waste. The core definition comes first, followed immediately by the crucial scope caveat, keeping the most decision-relevant information front-loaded.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a one-parameter, read-only list tool with no output schema, the description is complete: it explains what is returned (date, approximate headcount, source), what is excluded (all-cause layoffs), and the curated source. An agent has everything needed to call it correctly.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, and the single optional limit parameter is fully documented in the schema with its default and bounds. The description adds no parameter-specific meaning beyond the schema, so the baseline 3 applies.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description states a specific resource (recent layoffs where the company cited AI), names the data source (AI-Safe Careers' curated tracker), and explicitly distinguishes it from all-cause layoff datasets. Sibling tools (get_ai_exposure, search_occupations) cover different domains, so the scope is clearly separable.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description makes clear when this tool applies: only company-cited AI layoffs, with all-cause layoffs tracked separately and never included. It does not name sibling tools or give explicit when-not-to-use guidance beyond excluding all-cause datasets, so it falls short of a 5.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Find occupations whose titles match a search, with each one's AI Exposure Score, band and page URL. Use it when a job title is ambiguous, then call get_ai_exposure with the exact occupation.
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | How many matches to return (default 8). | |
| query | Yes | Words from a job title. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The annotations already declare readOnlyHint, idempotentHint, and closed-world behavior, so the safety profile is covered structurally. The description adds the otherwise-unknown return contents (score, band, URL), which is genuinely useful context beyond annotations. It doesn't address empty-result behavior or ranking/ordering, leaving a small gap.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two tight sentences: the first front-loads what is returned, the second front-loads the usage condition and the chained call. Every clause carries information and there is no boilerplate.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With no output schema and a simple two-parameter read, the description covers the essentials: what it returns and how to chain into get_ai_exposure. Minor absence of guidance on empty results or how matches are ranked keeps it just short of complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, with both query and limit fully documented including bounds and default. The description only implies title-matching semantics, adding nothing the schema doesn't already say. Baseline 3 applies.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb and resource ('Find occupations whose titles match a search') and enumerates what comes back (AI Exposure Score, band, page URL). It also distinguishes itself from the sibling get_ai_exposure by positioning itself as the lookup step that feeds it.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Gives an explicit trigger ('when a job title is ambiguous') and names the follow-up tool with the exact handoff ('call get_ai_exposure with the exact occupation'). No explicit when-not-to-use or note about skipping the search for exact titles, so not a full 5.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
get_ai_exposurerecent_ai_attributed_layoffssearch_occupationsSource-cited register of layoffs linked to AI, plus a live 0-100 AI job-loss index. No key.
Task-by-task AI exposure assessment of a job (Spain): 0-100 index, task map, weekly plan.
Sourced, task-level facts on how AI changes jobs, for agents answering 'what should I do?'
Workplace safety citation counts, penalties and 29 CFR standards data for AI assistants
Add one secure layer between your agents and this server.