Plan reads from a task

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

Beyond the readOnly/openWorld annotations, it discloses that the planner is deterministic with no model call, that no file contents are returned, that build output and DB/log artifacts are excluded from ranking, the cost profile (~20 ms per bounded search, 1-3 searches), and that it is batch-safe. These are non-obvious traits an agent could not infer from the annotations. Note: 'deterministic planner' sits with idempotentHint=false, but that is explained by recent-file inputs changing over time, not a 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?

Front-loaded with the mechanism, then labeled sections (Use when / Avoid when / Speed / Output / Batch-safe) that make it scannable. Slightly dense and longer than typical, but nearly every clause carries information an agent needs; only the 'deterministic planner, no model call' line overlaps the Speed 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?

There is no output schema, yet the description spells out the return shape (plan_read_batch with task, keywords, searched, candidates with relative_path/reason/hits, steps with tool/arguments/why, estimated_ms, notes) plus runtime cost. For a 4-parameter planning tool this covers everything needed to invoke and consume it 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 files_hint (ranked first, no search), include_search, and max_files are already fully documented in the schema. The description adds only loosely related detail ('bounded content + file-name searches (+recent files)') without mapping it to individual parameters, 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?

The description states a specific compound verb and output: extract keywords, run bounded searches, rank candidates, and return a ready project_read_batch call with an estimate. It explicitly distinguishes itself from project_read_batch (which it prepares a call for) and project_search (one-line lookups), so an agent can pick the right tool 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?

Contains explicit 'Use when' (unfamiliar area, unknown files, one call instead of guess-read-repeat) and 'Avoid when' (exact files known → project_read_batch; one-line lookup → project_search) clauses, each naming the alternative tool. This is the when/when-not/alternative pattern stated in full.

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