Return action classes, workflow action envelopes, runtime policy, and evidence.
| Name | Required | Description | Default |
|---|---|---|---|
| decision | No | ||
| risk_tier | No | ||
| workflow_id | No | ||
| action_class_id | No |
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Changes observed during successful MCP inspections. Dates show when Glama detected each change.
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden of behavioral disclosure. It only enumerates return content; it says nothing about the read-only nature, how the four optional parameters affect results, defaults, or null handling. For a tool without annotation coverage this is a significant, undisclosed behavioral gap.
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?
A single efficient sentence with the verb front-loaded and no wasted words. However, it is closer to under-specification than disciplined conciseness — a one-line list of jargon artifacts carries almost no usable substance.
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?
A four-parameter tool with 0% schema coverage, no annotations, and no usage guidance gets only a line listing return artifacts. The output schema exists, but the parameters and the semantics of the returned entities are entirely unexplained, so the definition is not adequate for confident invocation.
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 description coverage is 0%, and the description mentions none of the four parameters (decision, risk_tier, workflow_id, action_class_id). The description adds zero meaning beyond the bare parameter names, so the agent has no idea how to populate them or how they filter the returned artifacts.
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 uses a specific verb ('Return') and lists four concrete deliverables (action classes, workflow action envelopes, runtime policy, evidence). However, these artifact types are opaque domain jargon with no elaboration, and nothing distinguishes this from the ~80 similarly named recipes_agentic_* siblings that also deal with runtime artifacts (e.g., recipes_agentic_run_receipt_pack).
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?
The description offers zero guidance on when to use this pack versus its near-identically named siblings. No when/when-not conditions, no alternative tool names, no exclusions or prerequisites. Given the massive overlapping sibling catalog, an agent gets no routing help at all.
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.
Many tools return 'pack' artifacts with nearly identical descriptions, such as recipes_agentic_assurance_pack, recipes_agentic_posture_snapshot, and recipes_agentic_readiness_scorecard, or recipes_mcp_connector_intake_pack versus recipes_mcp_connector_trust_pack. Distinct domains like CVE lookup and playbooks are clear, but dozens of evidence/profile packs blur together and will cause misselection.
All names use the recipes_ prefix and snake_case, and most pack tools follow a [domain]_[topic]_pack pattern, which aids recognition. However, verbs are placed inconsistently and mixed with noun-only names: recipes_get, recipes_cve_get, recipes_mcp_server_get, recipes_refresh, and many pure 'pack' names.
Seventy-five tools is an extreme count for any MCP server, especially when the majority are highly specialized 'pack' endpoints with narrow outputs. The sheer number creates major selection overhead and makes the tool surface difficult for an agent to navigate reliably.
The server covers its apparent read-only scope thoroughly: recipe search/get, CVE lookup, playbook planning, MCP server catalog, upstream MCP introspection, and extensive evidence packs. There are no obvious dead ends, though the massive pack proliferation makes it harder for agents to know which tool to call.