experience_list

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

Annotations only declare readOnlyHint=true; the description adds the substantive behavioral rule that PROVENANCE determines citability, and names which values (cv_import, stated_in_chat, manual) may be cited versus agent_inferred, which may not. That is real safety-relevant context beyond the annotation. It does not mention volume/size of the bank or any pagination behavior, so it stops short of 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?

Three sentences, front-loaded with what is read, then the provenance rule that governs downstream use, then the cv_edit hand-off. It is dense but every clause carries information. Minor deduction for the shouty "PROVENANCE" and backticked identifiers, which slightly impede scanning.

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 does the work of describing the return structure: employments, their nested achievements, the `unplaced` bucket, per-achievement provenance, and the evidence_id field. That covers what an agent needs to consume the result. A note on whether achievements are ordered or how large the payload typically is would close the remaining gap.

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 tool takes zero parameters, so there is nothing for the description to disambiguate; baseline 4 applies. The description instead spends its words on output semantics, which is the right allocation.

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?

"Read the candidate's whole experience bank: every employment and the achievements attached to it, plus the ones attached to no place under `unplaced`" pairs a specific verb with a specific resource and the exact shape of what's returned. It is clearly distinguishable from the experience_add_*/update/remove siblings, which mutate, and from cv_get, which reads a rendered CV rather than the underlying bank.

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?

"Take evidence_id for cv_edit from here" explicitly routes the agent to the downstream sibling and states the precondition for using it. It does not, however, state when NOT to call this (e.g. when cv_context would be preferred, or whether a fuller profile read exists), so the exclusion side is only implicit.

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