ZEN SecDB

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare this as a safe read-only, idempotent operation. The description adds useful behavior beyond that: results are ordered by date, empty series should be returned when no history exists, and malformed CVE IDs should be corrected before calling.

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?

The description is front-loaded and clearly structured with headers and bullets. It is somewhat longer than necessary for a one-parameter tool, and the Outputs section is partly redundant because an output schema exists.

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 simple CVE history lookup, the description is complete: it covers purpose, usage, alternatives, input expectations, output shape, and empty-result behavior. No critical agent-facing information is missing.

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?

The schema has 100% coverage and already documents cve_id as a CVE-YYYY-NNNNN string. The description reinforces that format and adds validation behavior for malformed or incomplete IDs, plus guidance to call once per CVE when multiple are mentioned.

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 verb and resource: getting the historical EPSS time series for a specific CVE. It also explicitly distinguishes this from the sibling tool vulnerability_score by saying to use that tool for the current EPSS score.

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?

It gives concrete user intents that should trigger this tool, such as EPSS trend questions, spiking/dropping, and historical comparisons. It names an alternative tool for current-score questions and covers multi-CVE and malformed-CVE handling.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint/idempotentHint/destructiveHint=false and openWorldHint=false, so the safety profile is covered. The description adds value beyond that by documenting error behavior ('return a clear not-found error instead of guessing an alternative') and the shape of the response (summary + structured report). It stops short of noting rate limits or auth requirements, but for a read-only report runner this is solid.

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?

Front-loaded and clearly sectioned with headers, so an agent can scan it quickly. It is somewhat long, and the Outputs section partially duplicates the existing output schema, which is minor redundancy rather than noise.

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 an output schema present, the description needn't explain return values, yet it still summarizes them for chaining guidance. Combined with full schema coverage and rich annotations, nothing an agent needs to invoke this correctly is missing.

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%, so the schema already documents report_id, filters, and limit, and the description's Inputs section largely restates them. It does add one useful semantic point: report-specific filters and the instruction not to guess report IDs, directing the agent to the catalog. Baseline 3 is appropriate since the schema does the heavy lifting.

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 verb ('Run/Executes') and resource ('predefined report on ZEN SecDB public feed data') and enumerates the domains covered (CVEs, advisories, EPSS, weaknesses, etc.). It clearly separates this aggregate/analytics tool from the sibling lookup tools (vulnerability_info, vulnerability_search).

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?

Explicit when-to-use list (distributions, trends, top-N rankings, timeline summaries) plus an explicit when-not-to-use rule directing single-item requests to the dedicated lookup tools. It also names feed_report_catalog as the discovery route, which is the correct alternative for finding report IDs.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint=true, idempotentHint=true, destructiveHint=false, and openWorldHint=false, so the safety profile is covered. The description adds useful context beyond that: no input is required, the call is purely a catalog lookup, and results must be treated as the authoritative set of valid IDs. It stops short of describing ordering or size/caching 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?

Front-loaded with a one-line summary and clean headers, so an agent can skim it quickly. It is somewhat longer than necessary for a zero-parameter discovery tool, since the 'discovers reports for feed_report' point is repeated in the purpose, usage, and LLM guidelines sections.

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?

Everything needed to call it correctly is present: no inputs, the role relative to `feed_report`, and a description of the returned catalog fields. Although an output schema exists, the description still names the key return fields (report_id, title, description, input_schema), which helps the agent chain into `feed_report`.

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?

The tool takes zero parameters, so the baseline is 4. The description correctly and explicitly states 'This tool does not require any input', matching the empty schema with no contradiction.

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 ('List available public ZEN SecDB feed reports') and immediately frames the tool as a discovery counterpart to `feed_report`. An agent can distinguish it from every sibling without opening the schema.

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 when-to-use list (discover report IDs, inspect filters, unsure which report matches) and names the alternative tool `feed_report` with the ordering rule 'use before feed_report whenever the correct report ID is not already known'. It also states the negative constraint not to invent report IDs or parameters.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnly, idempotent, non-destructive, closed-world behavior, so the safety profile is covered. The description adds genuinely new behavioral context: results are based exclusively on the user-supplied package list, the summary is pre-rendered Markdown safe to show directly, and the agent is warned never to guess vulnerability status. It stops short of stating failure modes (e.g., unsupported os/version handling) or limits.

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?

Heavily structured with front-loaded headers and zero wasted sentences; the command snippets and example output directly serve the tool's strict input format requirement. Minor restatement of schema field text is offset by the otherwise high signal density.

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?

An output schema exists, yet the description still labels the two return fields (`report`, `summary`) and their intended use cases. Combined with input format guidance and prerequisite routing to `linux_os`, an agent has everything needed to invoke this tool 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%, so the baseline is 3. The description goes beyond the schema by giving the exact generation commands per distro family (rpm -qa, dpkg-query, apk list) and a sample package list, which is real added meaning for the `packages` parameter that the schema does not supply.

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 (audit), a specific resource (installed Linux packages), and the backing data source (SecDB). It is clearly distinguishable from the sibling `purl_audit` (which would audit by package URL) and from `linux_os` (which only enumerates supported OS/version pairs).

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?

Contains an explicit 'When to use this tool' section with three concrete triggering conditions, plus an explicit routing rule: if `os`/`version` are unknown, call `linux_os` first. This is a direct when/when-not/alternative disclosure.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint, idempotentHint, openWorldHint=false and destructiveHint=false, so the safety profile is covered. The description adds genuinely useful context beyond that: it is a static reference lookup requiring no input, returns a fixed OS/version array, and should never be guessed at. It does not mention caching or freshness of the list, which is the only real 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?

The opening sentence is well front-loaded, but the body is padded: the validation guidance is repeated across 'When to use this tool' and 'LLM usage guidelines,' and a full JSON example is shown even though an output schema already exists. The headers add structure but the content is roughly twice the necessary length.

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 zero-param lookup with an output schema, the description is more than sufficient: it explains the call ordering relative to `linux_audit` and the validity-checking role. The example output partially duplicates the output schema, but nothing an agent needs to invoke it correctly is missing.

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?

The tool takes zero parameters, so the baseline is 4. The description confirms 'This tool does not require any input,' which is consistent with the empty schema, though it adds no further semantic detail beyond that.

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 first sentence states a specific verb and resource ('List supported Linux operating systems and their corresponding versions') and explicitly ties it to the `linux_audit` sibling tool. An agent can distinguish it from every other sibling in the list without opening a schema.

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?

It gives explicit when-to-use conditions (user doesn't know supported OS names/versions, ambiguous OS info, need to validate before an audit) and states a sequencing rule: call this before `linux_audit` when parameters are uncertain. The 'do not guess' instruction further routes behavior.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint=true, idempotentHint=true, and destructiveHint=false, so the safety profile is covered. The description still adds real behavior beyond that: results depend exclusively on the supplied package list, unsupported ecosystems are silently ignored, and the summary/report outputs are distinguished. It doesn't disclose rate limits or failure modes, so a 4 rather than 5.

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?

Content is front-loaded with a one-line purpose followed by cleanly separated What/When/Ecosystems/Inputs/Outputs/Guidelines sections. It is long, and the ecosystem table plus manifest list are verbose, but each block carries actionable information rather than filler.

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 single-parameter, annotated, read-only tool with an output schema, the definition covers everything an agent needs: when to use it, supported inputs, and what report vs summary contain. Nothing required for correct invocation is missing.

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 'purls' parameter, so baseline is 3. The description adds meaningful semantics beyond the schema — per-ecosystem PURL syntax examples and a mapping of which manifest files (package.json, requirements.txt, Cargo.lock, etc.) each should be generated from, which genuinely helps correct invocation.

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 ('Perform a software package vulnerability audit') and pins the scope to PURL-identified packages backed by SecDB, which clearly separates it from OS-level or feed-oriented siblings like linux_audit and feed_report. An agent can identify what it does and its input domain without opening the schema.

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 'When to use this tool' section gives three concrete triggering conditions (dependency CVEs, advisory impact, patching needs) and even an imperative 'always call this tool' rather than guessing. It lacks explicit 'when not to use' or named alternatives among siblings, 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.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnly, idempotent, non-destructive, closed-world behavior, so the safety profile is covered. The description adds value beyond that by disclosing the data provenance (external exploit/scanning/social/MISP feeds), the shape of the result set, and the constraint that only valid enum values are accepted. It omits anything about result volume, pagination, or freshness of the aggregated feeds.

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?

The opening line is front-loaded and the headed sections are easy to scan. The 'Inputs' section, however, largely restates parameters the schema already documents at 100% coverage, which is redundant padding in an otherwise tight document.

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?

There is no output schema, and the description compensates by documenting the returned array and its fields (cve_id, status, category, reference, details). Combined with input semantics and usage triggers, an agent has everything needed to select and call the tool 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 description coverage is 100%, so the baseline is 3. The description rises above it by giving selection semantics rather than restating types: use cve_id for a targeted vulnerability, query for broad/exploratory searches, and only valid enum values for category/status. That is genuinely decision-useful guidance the schema alone does not convey.

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 verb and resource ('Search real-world vulnerability sightings in SecDB') and enumerates the concrete data it covers (Exploit-DB, Metasploit, Nessus/OpenVAS plugins, advisories, social, MISP). It is clear on its own, but never explicitly differentiates itself from close siblings like vulnerability_search or vulnerability_info, so an agent must infer the boundary.

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?

It provides strong contextual guidance via example user questions ('Is this CVE exploited in the wild?', 'Does Nessus have a plugin?') and an explicit preference statement ('Use this tool instead of assuming exploitation, PoCs, or plugin availability'). It does not, however, name the alternative sibling tools or state when *not* to use it, which keeps it 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.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already cover the safety profile (readOnlyHint, idempotentHint, non-destructive, closed-world), and the description adds real behavioral context: the tool cannot be called meaningfully without two user-supplied context values, and the summary field should be used to explain the decision. It does not discuss failure modes for unknown CVEs or rate limits, so it stops short of a 5.

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?

Header-delimited and front-loaded, so an agent can scan purpose, triggers, inputs, and outputs quickly. However, the 'Outputs' section enumerates fields that an output schema already provides, and the LLM usage bullets partially duplicate the When-to-use section.

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 three-parameter, deterministic calculator with full schema coverage, an output schema, and rich annotations, nothing an agent needs is missing: purpose, trigger conditions, input provenance requirements, and downstream composition with sibling tools are all present.

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%, so the baseline is 3, but the description adds the cve_id format (CVE-YYYY-NNNNN) that the schema omits and reinforces that M/S/E and M/A/I values are user-context inputs rather than derivable. The mission/well-being value explanations largely restate the schema, limiting the gain.

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 ('Compute CISA SSVC ... for a CVE') and names the exact decision categories returned (Track, Track*, Attend, Act). It clearly distinguishes itself from siblings like vulnerability_score and sightings_search, which are explicitly called out as complements rather than alternatives.

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?

Includes a dedicated 'When to use this tool' section listing three concrete user intents (remediation urgency, CISA SSVC priority, structured decision model) plus explicit LLM instructions to always collect mission_prevalence and public_well_being_impact before calling and never guess. It also names sibling tools to combine with.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and openWorldHint=false, so the safety profile is covered. The description adds genuine context beyond that: the SecDB data source, the structured Markdown return format, which fields may be absent ('if present', 'if included in the record'), and an explicit anti-hallucination instruction. It does not say what happens for a malformed or unknown CVE ID.

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?

Header-based structure is scannable and the purpose is front-loaded, but the field inventory is duplicated almost verbatim under 'What this tool does' and 'Outputs,' and the 'LLM usage guidelines' section restates the multi-CVE rule already given above. Roughly a third of the text 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 carries the return-value burden and does so thoroughly, listing every field family returned and the Markdown format. The only real gap is error/not-found behavior for an invalid CVE ID, which matters for a lookup tool with a single required key.

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 the schema already documents cve_id with the same CVE-YYYY-NNNNN format string the description repeats. The description adds no syntax, validation, or normalization guidance 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?

States a specific verb and resource ('Get detailed information about a specific CVE') and scopes itself explicitly to 'deep inspection of a single vulnerability,' which implicitly separates it from the multi-CVE siblings like vulnerability_search. The enumerated contents (CVSS, EPSS, advisories, CWE) make the resource concrete.

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?

Provides explicit when-to-use trigger phrases, states the multi-CVE handling rule ('call the tool once per CVE'), and names three complementary siblings (vulnerability_score, sightings_search, ssvc_calculator) with the condition for combining them. Nothing about selection is left to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint=true, idempotentHint=true, destructiveHint=false and openWorldHint=false, so the safety profile is covered. The description adds real behavioral value beyond that: it explains what cvss_explain contains, instructs the agent never to guess values, and says to call once per CVE for multi-CVE requests. It does not discuss caching, rate limits, or behavior on unknown/invalid CVEs, which keeps it short of a 5.

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?

Content is front-loaded and clearly sectioned, but it is long for a one-parameter tool and repeats the schema in an 'Inputs' section and the output schema in an 'Outputs' section, neither of which earns its place. The illustrative CVSS vector example is useful but disproportionate.

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 annotations covering safety/idempotency and an output schema present, the description already does more than required: it lists returned fields, explains the one non-obvious field (cvss_explain), and gives usage guidance. An agent has everything needed to call it correctly and interpret the result.

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 there is a single required parameter, so the schema already documents cve_id fully. The description restates the same CVE-YYYY-NNNNN format without adding validation rules, error behavior, or format nuances beyond the schema. 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+resource (get CVSS and current EPSS score for a specific CVE) and enumerates exactly what the risk snapshot contains. It also names the sibling that handles the adjacent case (epss_timeseries for historical trends), so an agent can differentiate without opening the schema.

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?

Provides an explicit 'When to use this tool' section with four concrete user phrasings, names the exclusion condition for EPSS historical trends, and routes to related tools (sightings_search, ssvc_calculator) for fuller assessments. Nothing about selection is left to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare read-only, idempotent, non-destructive, and closed-world behavior. The description adds substantial behavioral context beyond that: the exact searchable fields, supported Lucene syntax, default AND operator, result formatting as Markdown, and the presence of a search summary.

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?

The definition is long but well structured with headings and front-loaded purpose, making it easy to scan. Some sections, particularly Inputs and Outputs, repeat what the schema and output schema already convey, so it is not maximally concise.

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 search tool with rich annotations and an output schema, the description is highly complete: it explains scope, syntax, examples, output format, and usage guidance. An agent has everything needed to call and interpret the tool 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 there is only one parameter, but the description dramatically enriches its meaning by explaining Lucene syntax, listing searchable fields with examples, and showing concrete query patterns. This goes far beyond the schema's brief parameter note.

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 verb and resource — performing a full-text vulnerability search in SecDB — and enumerates the exact content types covered. It further distinguishes its exploratory search role from sibling analysis tools by naming vulnerability_score, epss_timeseries, and sightings_search as complementary functions.

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 gives clear when-to-use guidance with concrete user request examples such as looking up a CVE, advisory, exploit, or product, and labels the tool for exploratory or broad discovery. It does not explicitly name a when-not or a direct alternative for detailed single-record lookups, though it does route deeper analysis to siblings.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.