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

Annotations already declare readOnlyHint and idempotentHint, so the safety profile is covered. The description adds genuine context beyond them: the precedence rule for which file is read, that the path is returned alongside the content, and the external-SEARXNG_URL caveat. The return shape is left to the existing output schema.

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?

One front-loaded sentence that leads with the action and resource, with no wasted framing. Minor clutter from the inline 'CONCEPT:SR-KG.compute.embedded-instance' tag, which adds nothing for an agent deciding how to call the tool.

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?

No parameters, an output schema covering the return value, and the description supplies the resolution logic and the external-URL caveat. Complete enough to call correctly; the only residual gap is that it does not explicitly point to searxng_set_settings as the modification counterpart.

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?

Zero parameters, so the baseline of 4 applies; there are no parameters whose semantics need explaining.

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 (read) and resource (the embedded SearXNG instance's active settings.yml), and spells out the resolution order (user override at $XDG_CONFIG_HOME/searxng-mcp/settings.yml, else packaged default) plus the returned path. This clearly distinguishes it from siblings searxng_set_settings and searxng_reset_settings.

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 clause 'Informational only when an external SEARXNG_URL is configured' tells the agent when the result is not meaningful, which is real routing guidance. It does not explicitly name the alternative (searxng_set_settings) for modification, so it stops short of full when/when-not/alternative framing.

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?

No annotations exist, so the description bears the full burden. It does disclose valuable behaviour: best-effort semantics with 'ingested' being None when no engine is reachable, and the write shape (:Document, :SearchQuery, :SearchEngine nodes with :resultOf/:fromEngine links). It omits whether repeated ingests dedupe/idempotent, whether it mutates an existing graph destructively, permission requirements, and rate 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?

Front-loaded and short, but the body is broken by hard line breaks mid-sentence, and the trailing 'CONCEPT:AU-KG.ingest.enterprise-source-extractor' is internal taxonomy jargon that costs tokens without helping an agent decide or invoke. Roughly one of three sentences does no work.

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 five-parameter tool with a full output schema, the description covers purpose, the write target, and the failure mode (`ingested` is None), which is enough to call it correctly; return-value detail is rightly left to the output schema. The remaining gap is graph-side side effects and idempotency on repeat calls.

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 all five parameters carry their own descriptions, so the schema does the heavy lifting. The description adds no parameter-level meaning (e.g. engine naming conventions, category values, paging interaction with ingestion), so 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 compound verb plus resource: run a SearXNG search AND ingest results into epistemic-graph, naming the target graph and the node/link types created. That ingestion clause is exactly what separates it from searxng_search_app and web_search, without needing to name them.

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 'Wire-First: performs the search, then pushes each result into the knowledge graph' line implies the selection condition (you want results persisted, not just returned), but never states when to prefer this over searxng_search_app or web_search, nor any prerequisites such as an existing graph connection.

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 disclose that this is destructive, idempotent, and not read-only. The description adds useful behavioral context beyond that: it identifies exactly what is destroyed (the embedded user override settings.yml) and when the packaged default takes effect (on the next spawn).

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 a single, front-loaded sentence with no wasted words. It states the action, target file, and consequence immediately.

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?

The tool is simple, the input schema fully documents the only parameter, annotations cover safety traits, and an output schema exists. The description supplies the key behavioral effect, though it could explicitly mention the immediate-stop behavior governed by apply_now for a fully self-contained description.

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 single apply_now parameter is fully documented in the schema itself. The description does not add further parameter semantics, so the baseline of 3 is appropriate when the schema carries this responsibility.

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 uses a specific verb ("Delete") and resource ("EMBEDDED SearXNG user override settings.yml") and states the resulting effect (reverting to the packaged default on the next spawn). This clearly distinguishes it from read/write siblings like get_settings and set_settings, though it does not explicitly route to them.

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 phrase "reverting to the packaged default" implies this is used when the user wants to discard custom SearXNG overrides, but the description does not explicitly say when to use this versus searxng_set_settings or searxng_get_settings. No alternatives or exclusions are named.

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?

With no annotations, the description carries the full burden and does well: it discloses the backing mechanism (web_search via the host-mediated MCP Apps bridge) and the capability boundaries of the launched app. It stops short of stating what launching twice does or how app/session lifetime behaves, which are the remaining behavioral unknowns for a UI-launching tool.

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?

Three sentences, front-loaded with what is launched, followed by the selection rule and then the scope exclusions. Every clause contributes and nothing is repeated from the schema.

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-optional-parameter tool with an output schema, the description covers purpose, mechanism, and boundaries adequately. It is slightly incomplete on app lifecycle semantics (what a second launch yields, whether state persists), which an agent may need to call it confidently in a multi-turn session.

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 'query' parameter is already documented in the schema as an initial query run on load. The description adds no further parameter semantics (e.g., how an empty query behaves in the UI), 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 ('Launch') and a precisely scoped resource ('interactive SearXNG search-UI app'), then enumerates its components (query box, category filters, clickable results list). It explicitly distinguishes itself from web_search and from browser automation, so an agent can separate it from every sibling 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?

Gives an explicit selection rule: use when a human should see and interact with results visually rather than read raw JSON, which maps directly onto choosing this over web_search. It also pre-empts the two most likely misuses by stating it does not drive a browser or render arbitrary web pages.

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?

With annotations only declaring readOnlyHint=false and destructiveHint=false, the description carries real added weight: it discloses that the packaged default is never modified, that the override file is created on first use, and that the running instance is stopped so the next search respawns. It does not say what happens to in-flight requests when the instance is stopped, or what happens on invalid JSON, which keeps it from 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?

A single dense sentence that front-loads the merge action and then the restart consequence. No wasted sentences. Mild clutter from reST double-backtick markup, but nothing that impedes parsing.

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?

Output schema exists, so return values need no explanation, and all three parameters are documented. The description covers the mutation's blast radius and the embedded-vs-packaged distinction. It is only missing error behavior and whether prior override content is recoverable after a wholesale replace.

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 schema already documents all three parameters in detail. The description adds only the default `apply_now=true` behavior and the deep-merge semantics for `overrides_json`. 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?

States a specific verb and resource chain: deep-merge `overrides_json` into the embedded user override settings.yml and restart the embedded instance. That is far more than a restatement of the name. It does not explicitly contrast itself with siblings like `searxng_get_settings` or `searxng_reset_settings`, so sibling differentiation is inferred rather than stated.

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 well enough that an agent knows to call this to change settings, and it flags the `apply_now=false` path. However it never says when to prefer `searxng_reset_settings` over merging, or whether to read current settings with `searxng_get_settings` first. No explicit when-not guidance exists.

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 and idempotentHint=true, so the safety profile is covered. The description adds only that the backing service is a privacy-respecting SearXNG instance – no mention of rate limits, auth, result volume, or how 'privacy-respecting' affects behavior, so the added value is modest.

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 efficient sentence with the purpose front-loaded and no filler. It is perhaps overly terse for a tool with five parameters and multiple sibling alternatives, but nothing is wasted.

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 five parameters and an output schema present, the description covers the core action but omits anything about result shape, pagination limits, or how to choose engines/categories. Adequate but with clear gaps given the sibling set.

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 all five parameters (query, pageno, engines, language, categories) are documented in the schema itself. The description adds no syntax or format detail beyond that, 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 ('Perform a web search') and resource, plus the backend ('SearXNG metasearch instance'). It does not distinguish this from siblings like searxng_search_app or searxng_ingest_search, leaving the agent to infer which search variant to pick.

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?

No when-to-use guidance, no prerequisites, and no mention of when the sibling search tools (searxng_search_app, searxng_ingest_search) would be preferable. The agent must guess based on name alone.

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