collect_messages

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

Goes well beyond the readOnly/destructive/openWorld annotations: full observations are persisted as protected JSONL with no automatic deletion, paging is capped at 16KiB, cursor must be followed locally, resume continues unfinished Telegram work, and coverage caveats (preview-only search, unvisited scopes are not empty history) are disclosed. The untrusted-data warning is also valuable behavioral context.

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?

Telegraphic and dense, with compressed fragments like 'Default50/max100 per <=16KiB page, not collection cap' that cost readability. Most sentences carry real information, but the run-on packing and missing transitions hurt front-loading.

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 an 8-parameter tool with no output schema and no schema descriptions, the description supplies most of what an agent needs: scope, pagination, persistence, resume, and coverage limitations. Gaps remain around timeZone, the cursor's concrete format, and how queries interact with since/until.

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?

With 0% schema description coverage the description must compensate, and it largely does: chatIds omission semantics, since/until as ISO instants with Z/offset and until exclusive, source:local, limit default/max meaning per page rather than a collection cap, and cursor as a follow-next token. timeZone and the exact query-matching behavior are left unexplained.

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 (collect) plus resource (messages) and precise scope: a time interval or literal queries across allowed chats via a leased browser, or local history with source:local. It distinguishes the two operating modes clearly, though it never names the sibling tools (search_messages, get_history) it could be confused with.

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 real selection condition for source:local (no browser) versus the default Telegram path, and states that omitted chatIds means all current grants. It also notes resume semantics, but offers no explicit when-not or named alternative tools.

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