Skip to main content
Glama
Sanflow10

adversary-gate

by Sanflow10
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Local

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
ADVERSARY_GATE_POLICYNoGate flags (`--sandbox bwrap`, `--triage jev`, floors). An agent that can pass `--coverage-floor 0` grades itself. Evidence flags here are refused.
ADVERSARY_GATE_PYTHONNoThe interpreter's site-packages are harness too — a plugin installed there runs inside pytest. Use one the agent cannot write to.
ADVERSARY_GATE_BASE_REFNoThe baseline is the oracle: an agent could commit a rewritten test and name that commit as the base. Unpinned, the answer says "baseline_chosen_by": "agent".

Instructions

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

This server publishes no instructions, or was last inspected before Glama recorded them.

Capabilities

Features and capabilities supported by this server

Protocol revision2025-11-25

CapabilityDetails
tools
{
  "listChanged": false
}
prompts
{
  "listChanged": false
}
resources
{
  "subscribe": false,
  "listChanged": false
}

Tools

Functions exposed to the LLM to take actions

NameDescription
verify_repoA

Judge the working tree of repo against a baseline, by execution.

claims: pytest node ids (``path::test``) to verify; empty -> discovered from
the tests that executed a changed line. pytest_args: test paths or node ids
the coverage run collects (default: the whole suite); options are refused.
base_ref: used only when the operator did not pin ADVERSARY_GATE_BASE_REF.
gate_policyB

What this server enforces. Read-only: the agent cannot change it.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

B3.4/5.0

Scored across 2 tools

Disambiguation5/5

verify_repo executes tests and judges the working tree, while gate_policy returns read-only enforcement information. These purposes are completely distinct with no overlap, so an agent can easily select the right tool.

Naming Consistency3/5

Both names use snake_case, but verify_repo follows a clear verb_noun pattern while gate_policy is a bare noun phrase. This mixes action-oriented and resource-oriented conventions, though both remain readable.

Tool Count3/5

With only two tools, the surface feels thin even for a focused verification server. The tools are well-chosen but leave no room for auxiliary operations like baseline introspection.

Completeness4/5

The core verification and policy-reading operations are present, covering the main lifecycle for an adversary gate. Minor gaps exist, such as retrieving baseline details or historical gate results, but agents can work around them.

Maintenance

ActivityMaintained
ResponsivenessNo issues

Resources