Answers on Colorado corrections, jail and state agency policy, quoting the public documents used.
Scored across 1 tool
There is only one tool, so an agent cannot misselect between competing options. Its scope (answering policy questions with cited excerpts) is stated clearly enough to route queries to it unambiguously.
policy_rules_ask is snake_case and descriptive, but the verb is placed last rather than following the conventional verb_noun pattern (e.g. ask_policy_rules). With a single tool there is no set-level consistency to verify, so it lands just below full marks.
A single tool is thin for a domain covering multiple jurisdictions (state regulations, DOC regs, county jail policies). It earns its place, but one catch-all endpoint forces every interaction through the same surface.
The core retrieval-and-cite flow is covered, but there is no way to list available documents/jurisdictions, filter by source, or fetch a specific regulation section directly. These gaps are workable via the ask tool but limit structured lookups.
Answer a question from Policy Lens: Colorado Code of Regulations, Department of Corrections administrative regulations, county jail policies and state agency documents, with cited excerpts.
| Name | Required | Description | Default |
|---|---|---|---|
| language | No | Answer language (default en). | |
| question | Yes | Name the facility, agency, rule or topic. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint, idempotentHint, openWorldHint and destructiveHint=false, so safety and idempotency are covered. The description adds only that answers come 'with cited excerpts', which is useful return-shape context, but says nothing about accuracy limits, whether the answer is legal advice, latency, or what happens when no matching document exists.
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?
A single well-formed sentence with the action front-loaded and the source corpus as a scoping clause; nothing is redundant. It is dense but every element earns its place.
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 two-parameter, read-only, idempotent query tool with no output schema, the description supplies the domain scope and the key return characteristic (cited excerpts). What remains missing — behavior on no-match or out-of-domain questions — is minor at this complexity level.
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% and both parameters are documented there, including the 'Name the facility, agency, rule or topic' guidance and the en/es enum, so the baseline is 3. The description does not add any syntax, granularity, or phrasing advice beyond what the schema already states.
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 gives a specific verb ('Answer a question') plus the exact corpus scope: Colorado Code of Regulations, DOC administrative regulations, county jail policies, and state agency documents. That is far more informative than the title alone, though there are no siblings to be distinguished from, so it stops short of the top tier.
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 enumerated source corpus implicitly tells the agent what kinds of questions belong here (policy/regulation questions in Colorado corrections), but there is no explicit when-to-use or when-not-to-use statement, no mention of what happens for out-of-scope questions, and no alternatives are named.
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.
policy_rules_askColorado DORA license lookups, the SOMB provider directory and answers from the rules and standards.
Read-only public data on Georgia's prisons: population, facilities, deaths, contraband, statutes.
Free U.S. legal research: court opinions, statutes, state regulations and corrections policies.
Cited answers in Claude or ChatGPT only from your office's procedures; says so when not covered.
Add one secure layer between your agents and this server.