load_binary

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

Goes well beyond the annotations: annotations only say destructive/not-readOnly/idempotent, while the description states exactly what is lost (session names) and what happens on failure ('a refused load changes nothing'). That failure-atomicity detail is precisely the kind of behavior an agent cannot infer from hints.

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 short sentences, zero filler. The destructive side effect and the save_project prerequisite are front-loaded right after the identification clause, ordering the risk before the agent commits.

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 7-parameter destructive load with no output schema, the description covers identification, downstream effect ('the program all tools read'), return shape ('returning its summary'), data loss, and failure semantics. An agent has everything needed to invoke it safely.

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 every parameter including pdb, base, entry, regions, and entry_file_offset is already documented inline. The description only echoes the arch usage, adding no format or default details beyond the schema, so 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?

States a specific verb (load) and resource (PE/ELF/Mach-O file or raw bytes with arch), plus the crucial scope fact that it becomes 'the program all tools read'. This distinguishes it from load_project, reload, and fetch_pdb without needing their schemas.

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?

Explicitly tells the agent to call save_project first because session names are cleared, and signals the arch path for raw bytes. It doesn't cover when to prefer load_project or reload, so it stops short of full alternative routing, but the prerequisite guidance is concrete.

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