llms.txt for agents

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

Annotations already cover readOnly/openWorld/idempotent, so the safety profile is known. The description adds useful behavioral nuance beyond that: the HEAD-probe caveat for llms-full.txt existence and the fact that all attempted rungs are returned for diagnosability. This goes beyond what annotations express, though it stops short of describing edge cases like network failures or malformed llms.txt files.

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 two sentences with no filler. The first sentence front-loads the core operation and details what is parsed; the second gives routing guidance and a behavioral promise. Every clause earns its place, and the structure is easy to scan.

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, so the description shoulders the burden of explaining return content, and it does list the key parsed sections. It also prepares the agent for miss behavior. Minor gaps remain: the exact return shape is not described, and llms_search is not referenced as an alternative, but overall the description is sufficient for correct invocation in most scenarios.

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 parameters (site, refresh, task_context) are already well documented in the schema. The description adds no parameter-level semantics beyond what the schema states, which matches the baseline of 3 for high schema coverage.

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 leads with a specific verb and resource: 'Find and parse the llms.txt that covers a site or page URL', then enumerates exactly what gets parsed (title, summary, H2 sections, Optional section, llms-full.txt probe). It also explicitly names the sibling it is differentiated from ('Use first, before llms_page'), so an agent can tell them apart.

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 a clear when-to-use rule: 'Use first, before llms_page, whenever the user needs a product or library's own docs'. It also provides diagnostic context ('every rung tried is returned so a miss is diagnosable') that helps the agent decide whether to call a fallback. This is explicit, actionable guidance.

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?

The description discloses a useful fallback chain: Accept negotiation, markdown mirror, .md conventions, then bounded HTML-to-markdown conversion. This is meaningful behavioral context beyond the annotations, and it aligns with the read-only, idempotent, and non-destructive hints. It does not contradict the annotations.

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 tightly packed sentences: the first front-loads the primary function and conversion strategy, the second gives usage guidance. Every clause contributes, with no filler or repetition of schema details.

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 paginated markdown fetcher, the description adequately conveys the return type, pagination behavior, and conversion pipeline. It does not describe failure modes or exact response envelope (e.g., next_offset), but the schema and annotations cover parameter and safety aspects, making the definition reasonably 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%, so the schema already documents url, offset, max_chars, and task_context. The description adds only indirect context about pagination and bounded conversion, which slightly enriches offset/max_chars semantics but does not substantially go beyond what the schema provides.

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 clearly states a specific operation: 'Return one docs page as markdown, paginated' with a precise verb and resource. It also distinguishes itself from siblings by referencing URLs from llms_index and focusing on individual page retrieval rather than index discovery or 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?

It explicitly instructs when to use the tool: 'Use on the URLs llms_index returns, or on any docs URL you already hold.' This gives a clear invocation context, though it does not explicitly name alternative tools or state when not to use it, so it stops short of a full exclusionary guide.

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?

The annotations already establish read-only, idempotent, non-destructive behavior, so the description adds meaningful context: weighted matching across section headings (×3), link titles (×2), and notes (×1), plus on-demand indexing per site. This goes beyond the annotations and helps an agent predict how results are ranked.

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 three sentences with no wasted words. It front-loads the search action and resource, then adds the key behavioral details and the primary use case. The weighted scoring notation is compact but informative.

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?

Given no output schema, the description adequately states what is returned (matching doc links and covering index). The annotations cover safety, and the schema covers parameter constraints and the task_context rationale. Minor gaps include not describing the exact result format or ordering, but the overall picture is complete enough for correct invocation.

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 high at 80%, so the schema already documents site, query, max_bytes, and task_context. The description adds value by explaining the weighting behind query matching and reinforcing that site restricts to one site and is indexed on demand. It does not cover limit or max_bytes, but the schema handles those adequately.

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 opens with a specific verb and resource: 'Search the llms.txt indexes this server has parsed' and states the concrete output: 'matching docs links with their covering index.' It also distinguishes itself from the sibling tools by framing this as a way to jump to the right page instead of reading the whole index.

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 a clear use case: use it to jump straight to the right page of a docs set instead of reading the whole index. It also explains when to pass the site parameter ('search one site (it is indexed on demand)'). It does not explicitly name the sibling tools or state when not to use them, but the context is clear.

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