Skip to main content
Glama

Mnemom — Trust Ratings for AI Agents

by io.github.mnemom

preview_compose_protection_by_agent

Read-onlyIdempotent

Preview composed protection (dry run) — Composes the cascade against a hypothetical body at the agent layer and returns conflicts + the composed view. No DB writes. Used by the dashboard editor for live conflict markers.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
modeYesScreening mode for the protection pipeline. Required. `off` disables screening; `observe` records only; `nudge` warns; `enforce` blocks.
agent_idYesThe agent this card belongs to (e.g. `smolt-abc123`). Identifier only — never place an API key, a secret, an email address, or any other personal data in this field.
thresholdsNoRisk-score cutoffs, each in [0, 1] and ordered warn ≤ quarantine ≤ block. All three are required if this object is sent at all — omit the whole object to accept the composed defaults.
card_versionYesCard schema version. REQUIRED by the server-side validator. Current canonical value: `protection/2026-04-26`.
screen_surfacesNoWhich traffic surfaces are screened. Omit to accept the composed defaults.
trusted_sourcesNoSources exempt from screening. Enumerate specific hosts — wildcards are rejected, and a server-side deny-list (public LLM/DNS endpoints, 0.0.0.0/0, ::/0, link-local, multicast) is always applied.
protected_surfaceNoThe assets and operations this agent must protect. Omit to accept the composed default (empty surface).

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault
okYesTrue when composition succeeded (no blocking conflicts).
summaryYesOne-line human-readable summary of composition status.
full_reportNoOptional pointer to the full /v1 conflict report (method + path).
conflicts_countYesTotal number of conflicts detected (0 = none).
composition_validYesTrue when the composed card is valid.

TDQS

A4.4/5.0
Behavior5/5

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

Annotations already declare readOnlyHint=true, idempotentHint=true, and destructiveHint=false. The description adds valuable behavioral context beyond these: it explains the tool performs a dry run against a hypothetical body, returns conflicts plus the composed view, and explicitly confirms no database writes. This aligns with annotations and enriches understanding without contradicting them.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is two sentences, front-loaded with the core purpose ('Preview composed protection (dry run)') and immediately conveys key traits ('No DB writes', returns conflicts + composed view, dashboard editor use). Every word earns its place with no redundancy.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's complexity (7 params, nested objects, output schema), the description covers the high-level behavior, safety profile, and intended usage. It does not need to explain return values because an output schema exists. It could have elaborated on what a 'hypothetical body' is, but the annotations, schema, and output schema collectively fill the gaps, making it sufficiently complete for an agent.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the baseline is 3. The description does not add specific parameter-level semantics beyond what the schema provides, and it does mention 'hypothetical body' but does not map that to individual parameters. However, the schema descriptions are already rich, so no significant gap exists.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description uses a specific verb ('Preview composed protection (dry run)') and resource ('protection'), and clearly states the tool composes the cascade against a hypothetical body and returns conflicts + composed view. It also distinguishes itself from siblings by emphasizing 'No DB writes' and the dashboard editor use case, differentiating it from preview_compose_alignment_by_agent and put_protection_by_agent.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description specifies a clear use case: 'Used by the dashboard editor for live conflict markers.' It also implies a dry-run/no-persistence context with 'No DB writes,' which suggests when not to use it (i.e., when persistence is needed), though it does not explicitly name put_protection_by_agent as the alternative. This is clear context but lacks explicit exclusions.

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

All Mnemom — Trust Ratings for AI Agents toolsInspect
Try in Browser
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

A4.1/5.0
Disambiguation5/5

Each tool targets a distinct operation: identity claiming, lookup, reputation retrieval/badge, scanning, verification, alignment/protection management, and feedback. No significant overlap exists.

Naming Consistency5/5

All tools follow a consistent verb_noun pattern using snake_case (e.g., claim_agent, get_reputation, verify_scan). Even complex names like preview_compose_alignment_by_agent adhere to the pattern.

Tool Count4/5

With 16 tools, the set is slightly heavy but still well-scoped for the domain of AI agent trust ratings. Each tool serves a clear purpose, and no tool feels redundant.

Completeness4/5

The surface covers core workflows: agent identity, reputation, alignment/protection, scanning, verification, and feedback. Minor gaps like agent updates or deletion might exist, but the core lifecycle is complete.

Resources