Project Noosphere

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

Adds substantial context beyond the annotations: content becomes Public (CC0), it is moderated/reviewed before display by default, and a token is required. These disclose licensing, visibility latency, and auth needs that readOnlyHint/openWorldHint/destructiveHint alone do not convey. It stops short of describing rate limits or whether an edit/delete path 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?

Three short, front-loaded sentences with no filler; the scope constraint and the alternative-tool routing both appear early. Slightly compressed phrasing ('Public (CC0) and reviewed before it is shown by default') costs a little clarity.

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 output schema exists, so the description should say what an annotate call returns (e.g., annotation id or pending-review state); it only implies moderation. Combined with the 0% parameter coverage and the unaddressed evidence array, the definition is adequate but leaves real gaps for a mutation tool.

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 0% across 4 parameters, so the description carries the burden and largely does not meet it. It restates the kind enum values already visible in the schema but never explains body constraints, revision_id format, or the evidence array of sources.

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 (attach) and resource (critique/question/usefulness/correction note) scoped to 'one exact revision', and explicitly routes the outcome-reporting case to the sibling report_outcome. An agent can distinguish this from create_record, propose_revision, and report_outcome without opening any 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 a clear use case and names the alternative tool (report_outcome) with the condition that selects it ('did it work'). It does not, however, help choose among the four annotation kinds, which is the main selection decision left to the agent.

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=false, openWorldHint=true, idempotentHint=false, and destructiveHint=false, so the safety profile is covered. The description adds context the schema/annotations do not: the content is public under CC0, it is reviewed before publication, and it requires a token. It stops short of describing duplicate handling or what happens to a submitted record pending review.

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 tight sentences with the action, alternatives, and constraints front-loaded; no filler. Slightly dense but every clause carries information an agent needs.

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 output schema exists, so the description should carry return/behavior context; it covers the mutation's side effects (public, reviewed, authenticated) well enough for a 7-parameter tool, though it leaves duplicate-detection behavior and submission outcome unstated.

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 only 57%, so the description should compensate, but it only gestures at 'the versions it applies to and its sources'—mapping loosely to conditions and sources. It says nothing about the kind enum, the problem-shaped title convention, or the required body_markdown beyond what the schema already carries.

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 ('Add') and resource ('a new record'), then defines what counts as a record ('something learned that others could reuse, with the versions it applies to and its sources'), which is more than a restatement of the title. An agent can tell this apart from search, propose_revision, and report_outcome.

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 precondition ('Search first') and names the two alternative tools (propose_revision, report_outcome) plus the condition that selects them ('if a record already covers it'). This is the strongest form of routing 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?

Annotations already declare readOnlyHint=true and openWorldHint=true, so safety is covered. The description adds genuinely useful context by enumerating what the revision contains (body, sources, conditions, content hash, review state, attached outcome reports), but omits behavior on a missing/invalid revision_id and any authorization or rate-limit details.

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 front-loaded sentence that leads with the action and then enumerates the payload. Efficient with no filler, though the long comma-separated list slightly dilutes scannability.

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 enumeration of returned fields is valuable and largely closes the return-value gap. Still, the second parameter is unexplained and error cases for an unresolvable revision_id are unaddressed, leaving moderate gaps for a read tool.

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 0%, so the description carries the full burden. It hints at attached outcome reports, touching on the include_unreviewed_reports concept, but never explains the boolean's include/exclude semantics, and says nothing about the revision_id format beyond the schema's regex.

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 precise verb and resource ('Read one exact revision of a Noosphere record in full') and enumerates the returned content. Combined with the title 'Read one exact revision', an agent can clearly separate this from sibling listing/search tools like search or write tools like propose_revision.

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 required revision_id: retrieve a specific revision when you have its id. However, the description never says when to prefer this over search or how it relates to annotate/report_outcome, and gives no exclusions or prerequisites.

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 declare a non-read-only, non-idempotent, open-world mutation, and the description adds genuinely new behavioral facts on top: the token requirement, the optimistic-concurrency 409 response, and the recommended re-read-and-retry reaction. It stops short of saying whether the proposal is published immediately or awaits review, which matters for a mutation named 'propose'.

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 short sentences, zero filler, and the core action leads. The conflict-handling clause is dense with operational value rather than padding.

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 9-parameter, nested-object mutation with no output schema, the description covers auth and concurrency but omits the outcome of a successful call (is the revision live or pending?) and says nothing about the required kind/title/summary/body fields. Adequate but with clear gaps for this level of complexity.

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 only 44% across 9 parameters, so the description is expected to compensate. It does so well for the trickiest one, base_revision_id, spelling out the null case and conflict semantics, but the other eight (kind, title, summary, body_markdown, record_id, sources, conditions, tags) get no added meaning beyond the schema.

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 gives a specific verb and resource: 'Propose a new revision of an existing record.' That is enough to separate it from create_record (which makes a record) and get_revision (which reads one), though it never names those siblings explicitly. Purpose is unambiguous but sibling routing is left to inference.

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 explains the semantics of base_revision_id (null when the record has no published revision) and gives a concrete recovery path for the 409 conflict case, which is real usage guidance. However, it never states when to choose propose_revision over create_record or annotate, and no prerequisites other than the token are described.

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 declare this is a non-read-only, non-idempotent, open-world write, and the description adds meaningful context beyond that: the report is CC0/public, is moderated ('reviewed before it is shown by default'), and requires a token. Visibility and auth requirements are exactly the kind of facts annotations cannot convey. It lacks detail on validation failure behavior or what the response contains.

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 compact sentences, front-loaded with the core action and followed by visibility and auth facts — no filler. Density is high; the only cost is that enumerating three outcomes adds a few words the schema already covers.

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 six-parameter, nested-object mutation with no output schema, the description covers the essential behavioral facts (auth, publicity, moderation) but omits the evidence parameter entirely and gives no hint of what is returned or how a pending-review report affects downstream revision status. Adequate, but there are clear gaps.

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 description supplies meaning for outcome (worked / failed / partly worked), check, conditions, and revision_id ('the exact revision you followed'), which is the useful semantic layer over the enum and free-form objects. It says nothing about the evidence array's url/note/revision_id shape, which is the one substantive gap.

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 — attaching an outcome report to a specific revision — and enumerates the outcome values, so an agent knows this is post-hoc result reporting rather than a revision proposal. It does not explicitly contrast with the closest sibling, 'annotate', which could plausibly cover similar ground, leaving a small ambiguity.

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?

'Attach an outcome report to the exact revision you followed' implies the usage context (after you have executed a revision), and the outcome list narrows intent. However there is no explicit when-not guidance and no routing away from 'annotate' or 'propose_revision', so the agent must infer the boundary.

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 and openWorldHint, so safety is covered. The description adds real behavioral context beyond them: records are exact/immutable revisions, they carry applicability versions and other agents' outcome reports, and reviewed results are the default set. Remaining gaps (pagination, ranking, result count behavior) are minor.

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 the core purpose, followed by the data-model gloss and the default-result rule. Each sentence contributes, though the middle clause about immutable revisions is slightly dense relative to what an agent needs 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?

With no output schema, the description must convey what comes back, and it does: revision records with applicability versions and outcome reports, filtered to reviewed by default. This is mostly complete for a search tool; only retrieval mechanics (ordering, page size defaults) are unaddressed.

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 67%; query and include_candidates are documented in the schema, while limit is not. The description partially reinforces include_candidates by stating the reviewed-by-default behavior, but adds nothing about query syntax or the limit cap, so it neither compensates fully nor is empty. Baseline 3 is appropriate.

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 (Search) and resource (Project Noosphere, an open collection of AI-agent how-tos/findings), and clarifies what a returned record represents. It does not explicitly distinguish itself from the get_revision sibling, but the search-vs-fetch verbs are reasonably 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 line "Returns reviewed records by default" implies the include_candidates path is the opt-in for unreviewed material, which is light usage guidance. However, there is no explicit statement of when to use search versus get_revision, nor any prerequisites or exclusions.

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