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

Annotations only flag it as a non-read-only, destructive, idempotent, closed-world write. The description adds what the annotations cannot: the value is sealed at rest, scrubbed from logs, never written to audit rows (audited by name only), and delivered to the process solely at invoke time. It also states the retry contract and the ANTHROPIC_API_KEY / BYOK semantics, which is substantial behavior beyond the structured fields.

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 action and the scope decision, and every clause carries information (invoke-time reach, sealing, precedence, retry safety, audit, return shape). It is one dense multi-clause block rather than a broken-out structure, which slightly raises scanning cost, but there is no filler to cut.

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 four-parameter mutation tool with no output schema, the description covers the auth requirements, the two scoping modes and their interaction, secret-value handling, idempotency, and even the return shape ({name, effect, updatedAt}). An agent has everything needed to call it correctly and interpret the result.

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 100%, so the baseline would be 3; the description goes further by explaining what 'loop' and 'org' actually select, the inheritance and override relationship between them, and the auth scope each requires — meaning the bare schema ('Organisation slug', 'Hosted Claude loop id (uuid)') does not carry. Name-pattern restrictions are already in the schema, so no credit there.

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 ('Create or replace a secret a hosted Claude loop runs with') and immediately scopes it: the value is reachable only as process.env.NAME at invoke time. It distinguishes the two target scopes (loop-level owner-only vs organisation-level admin-scoped, with loop overriding org) and is clearly separable from siblings like set_repo_secret or set_variable.

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?

Explicit routing guidance: pass 'loop' for the loop's own secret (owner only) or 'org' for an organisation-wide secret requiring an admin-scoped token, plus the precedence rule that a loop-level name overrides the organisation's. It does not contrast itself against sibling tools (e.g. when to use this rather than set_variable or set_repo_secret), which keeps it short of a 5.

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