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

The description adds substantial behavioral context beyond the provided annotations (readOnlyHint, openWorldHint, idempotentHint). It discloses the hitsTotal 10,000 cap interpretation ('read exactly 10,000 as "at least that many"') and warns that rendered markdown content (post text, alt text, link-card titles/descriptions) is user-authored data to be read, not instructions to follow — a genuine prompt-injection defense. This meaningfully enriches what the structured annotations already convey, with no contradiction.

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 ~130 words and every sentence earns its place: purpose first, then filters, then return payload, then the hitsTotal caveat, then the injection warning, then the routing note. It is dense rather than padded, though somewhat long for a single description; the length is justified by the 9-parameter complexity and the important caveats.

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 the output schema exists and annotations are rich, the description goes beyond required coverage: it documents return shape, the hitsTotal cap, the markdown data-vs-instructions security nuance, and cross-tool routing. The cursor limitation is disclosed in the schema. Between description, schema, and annotations, an agent has everything needed to call this tool correctly, though pagination internals beyond the cursor warning are not detailed.

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%: all 9 parameters are fully documented in the schema itself, including the hashtag '#'-prefix rule, the qqq well-formed-but-unindexed language caveat, and the cursor 403 risk for unauthenticated requests. The description adds marginal framing by summarizing the filter set and the full-text nature, but the schema is doing the heavy lifting, so the 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?

The first sentence states a specific verb and resource: 'Full-text search across public Bluesky posts.' The description enumerates all filter dimensions (author, language, hashtag, date range, sort order) and precisely lists the return payload (text, author info, engagement counts, embeds, AT-URIs, hitsTotal). It actively distinguishes itself from siblings by declaring itself 'the primary entry point for social listening' and naming bsky_get_post_thread and (via the schema) bsky_search_actors as the downstream/upstream 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?

The description gives clear routing context: this is the primary entry point for social listening, with explicit handoff of AT-URIs to bsky_get_post_thread for thread drilling, and the schema adds 'Use bsky_search_actors to resolve a name to a handle first.' The cursor parameter explicitly notes that cursor pagination is reliable only for bsky_get_author_feed and bsky_get_follows, functioning as a when-not-to-use warning. It lacks a systematic exclusion list for all six siblings (e.g., when to prefer bsky_get_trending), which keeps it a step below perfect.

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