Verify the structural integrity, agent binding, delegation continuity, and commitments of an Obelisk action-scoped run proof. Does not reveal or infer raw task data.
| Name | Required | Description | Default |
|---|---|---|---|
| run_proof | Yes | An obelisk-agent-run-proof-v1 object. | |
| expected_agent_id | No | Optional expected agt_ subject. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Although annotations are absent, the description carries the core behavioral contract: verification rather than exposure of task data, and the explicit commitment 'does not reveal or infer raw task data.' It does not disclose other behaviorly relevant details like return values, failure modes, or privacy assumptions beyond the data access claim; otherwise this would be a 4.
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?
Two short, front-loaded sentences state the full contract with zero redundancy. The main verb and the object are in the first sentence, followed by the key privacy constraint. No unnecessary elaboration.
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?
The tool has a nested object parameter and no output schema, but the description does not say what the return value is or how the verification result should be interpreted. The behavioral core is described, but the missing return semantics and usage context leave a significant gap in what an agent needs to confidently invoke and act on this tool.
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 input schema already provides 100% description coverage for run_proof and expected_agent_id. The description adds no connection between these parameters and the aspects of verification (e.g., how expected_agent_id is bound to the delegation binding check). The baseline 3 is appropriate because the schema does the heavy lifting.
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 resource (action-scoped run proof) and a precise verb (verify), enumerating the exact facets: structural integrity, agent binding, delegation continuity, commitments. It adds a clear negative guarantee about raw task data. It does not explicitly differentiate from siblings by name, but the focus on run proof verification strongly distinguishes it from tools like verify_token or get_agent_proof.
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?
No guidance on when to use this tool versus siblings or when not to. The information flows in one direction only: what the tool does. The note about not revealing raw task data hints at a usage context, but partners no alternative or decision rule, leaving the agent to infer coverage.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Every tool has a distinct concern: explaining ratings, reading gate status, fetching proofs, looking up org ratings, scanning URLs, checking transparency heads, and verifying proofs/tokens. Even the proof-related tools are cleanly separated into read (get_agent_proof) versus verify (verify_agent_run_proof).
Most tools follow a clear snake_case verb_noun pattern: explain_rating, scan_trust, get_org_rating, verify_token, verify_agent_run_proof. However, gate_status and transparency_head are noun phrases rather than command-style names, which is a minor inconsistency.
Eight tools is a well-scoped set for a read-only trust and verification server. Each tool covers a meaningful capability with no redundant or superfluous entries.
The core domain is covered well: rating explanation, rating retrieval, trust scanning, gate status, agent proofs, run op and token verification, and transparency anchoring. The main gaps are convenience functions like listing organizations or resolving proof IDs, but these are not essential to the primary workflow.