OPA MCP
OCI · DOCKER.IO/ORYGN/OPA-MCP · 2 COMPONENTS
Author, validate, debug, and explain OPA Rego policies through any MCP-compatible client.
Available components
Last scanned (3 hours ago)
Scores higher than 91% of OCI servers
Category Security
How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. How we score → Why this is hard to score →
Supply Chain Security0
- No malware scan is available for this kind of package: the supply-chain vendors we use do not cover it. This is a permanent gap in our coverage, not a finding about the package.Unverified
- Known CVEs could not be checked: this artifact ships no SBOM or dependency manifest, so there is no dependency list to read.Unverified
- Install-script risk not yet assessed.Unverified
- Dependency health could not be checked: this artifact ships no SBOM or dependency manifest, so there is no dependency list to read.Unverified
Provenance & Transparency48
- Source repository is publicly reachable at the declared URL. View diagnostics → Pass
- Provenance check failed: no build-provenance attestation is published. See how to fix → View diagnostics → Fail
- Clear OSI-approved license (MIT).Pass
- Actively maintained (last published 12 days ago).Pass
- Publishes a security disclosure policy (SECURITY.md).Pass
Schema Quality & AI Usability76
- 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).Pass
- AI-judged instruction clarity (excellent).Pass
- Context-footprint check failed: tool/resource definitions use about 17133 tokens (~311/item across 55 items; 52 tools + 3 resources), over budget; trim descriptions and params. See how to fix → Fail
- Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management100
- No destabilizing schema changes in the last 30 days.Pass
Tool Coverage100
- 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
- 100% of tool parameters carry a description.Pass
Tool Safety81
- No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
- 2 of 8 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation; "rego_eval" implies "eval" and declares no destructiveHint at all, which the MCP spec reads as destructive by default. See how to fix → Partial
- An AI judge read all 54 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
- Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
Unverified: 1 category
A category scored 0 because we could not verify it: a data source with nothing on this package, evidence we could not reach, or a check we could not run. We only credit what we can confirm.
Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.
- 2 Nov 26 0
- Provenance: unverified → fail ▼ security
- 1 Nov 26 0
- Provenance: fail → unverified ▼ security
- 28 Oct 26 0
- Stability: 0.97 → pass security
- 27 Oct 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 93 to 97. That category is still filling its 30-day observation window: 28 days of observed history at the previous scan, 29 at this one. The score rises as the window fills, whether or not the server changes.
- 26 Oct 26 0
- Provenance: unverified → fail ▼ security
- 25 Oct 26 +1
- Provenance: fail → unverified ▼ security
- 23 Oct 26 +1
- Package version: 0.9.0 → 0.9.1 functional
- 21 Oct 26 +1
- Provenance: unverified → fail ▼ security
Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.
Captured 5 Nov 2026 · Analysed oci/docker.io/orygn/opa-mcp@0.9.1
Provenance No attestation
The registry publishes no build provenance for this version, so there is nothing to verify.
| Result | No attestation |
|---|---|
| Ecosystem | oci |
| Reason | No attestation published |
Background: How many MCP packages publish verified provenance →
The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →
rego_test_multiroot Run Rego tests across multiple roots ~672
Run `opa test` once per root and aggregate results. Solves the package-conflict problem that occurs when `opa test .` is run on a repo with multiple independent package namespaces (OPA issue #4724). Two modes: `explicit` (supply root list with optional per-root `include` paths for shared libraries) and `scan` (auto-discover leaf test roots using the leaf rule -- a directory is a root only if it directly contains `*_test.rego` files and none of its eligible subdirectories do, preventing OPA's automatic recursion from double-running tests). Use `sharedPaths` in scan mode to add shared library directories to every root's invocation without including them in discovery. Coverage and threshold work per-root; `overallCoveragePct` is the mean across roots that have coverage data.
| Name | Type | Req | Description |
|---|---|---|---|
| coverage | boolean | – | Include per-line coverage data per root. Switches output to coverage-report mode: test record counts are not available, but `coverage`, `coveragePct`, and `overallCoveragePct` fields are populated. |
| ignorePatterns | array | – | Additional directory name patterns to skip during scan (e.g., ["vendor", "*.generated"]). Supports `*` wildcards. Only used with `scanDir`. |
| maxDepth | integer | – | Maximum directory depth to scan. Default: 10. Only used with `scanDir`. |
| maxRoots | integer | – | Maximum number of test roots allowed. Returns INVALID_INPUT if scan finds more. Default: 50. Only used with `scanDir`. |
| roots | array | – | Explicit list of test root directories. Use when roots are known upfront or when scan mode cannot determine the correct roots. Mutually exclusive with `scanDir`. |
| runPattern | string | – | Run only tests whose names match this regular expression (passed as `--run` to each root). |
| scanDir | string | – | Top-level directory to scan for test roots. Uses the leaf rule: a directory is a root only if it directly contains `*_test.rego` files and none of its eligible subdirectories do. Mutually exclusive w… |
| sharedPaths | array | – | Paths added to every root's `opa test` invocation and excluded from auto-discovery. Use for shared library directories that all roots import from. |
| threshold | number | – | Minimum coverage percentage required per root (0-100). Roots below threshold have `thresholdMet: false` in their result. Implicitly enables coverage-report output mode. |
| v0Compatible | boolean | – | Read the policy as Rego v0 (`--v0-compatible`), the syntax OPA used before 1.0: rules without `if`, partial sets as `deny[msg] { ... }`. Needed for a policy that has not been migrated, which OPA 1.x… |
| varValues | boolean | – | Include local variable bindings in trace output (`--var-values`). Only useful with `verbose: true`. |
| verbose | boolean | – | Emit per-test pass/fail details for each root. |
No output schema declared.
No examples provided.
rego_verify Formally verify a Rego policy rule ~410
Formally verify a property about a Rego rule using SMT solving (Microsoft Z3). Unlike testing, this checks ALL possible inputs and either proves the property holds or returns a concrete counterexample input that falsifies it. Supports equality, comparison, startswith, endswith, contains, and simple regex.match patterns (prefix: ^lit.*, suffix: .*lit$, exact: ^lit$, contains: .*lit.*, wildcard: .*). Complex regex patterns (character classes, quantifiers, alternation) return INCONCLUSIVE. Also reports INCONCLUSIVE for negation-as-failure (not), comprehensions, partial set and object rules (deny contains msg), functions, else chains, and any operand it cannot encode. A body that reads an absent field is undefined rather than true, so always_true holds only if the rule is also true for an empty input: a rule requiring input.x will be answered with the counterexample {}.
| Name | Type | Req | Description |
|---|---|---|---|
| kind | string | yes | Property to prove: always_true - rule is true for every possible input (finds inputs that violate this) never_true - rule is never true for any input (finds inputs that trigger it) satisfiab… |
| rule | string | yes | Name of the rule to verify (e.g. "allow", "deny"). |
| source | string | yes | Rego source to verify. |
| v0Compatible | boolean | – | Read the policy as Rego v0 (`--v0-compatible`), the syntax OPA used before 1.0: rules without `if`, partial sets as `deny[msg] { ... }`. Needed for a policy that has not been migrated, which OPA 1.x… |
No output schema declared.
No examples provided.
How do I install the OPA MCP server?
OPA MCP runs locally as a container image, launched with docker run --rm -i -e OPA_URL -e OPA_TOKEN -e GITHUB_TOKEN docker.io/orygn/opa-mcp:0.9.1. Ready-made configuration for Claude, Cursor, VS Code, Codex and 3 more is on this page, copied from each client's own documentation.
oci · docker.io/orygn/opa-mcp
claude mcp add orygnscode-opa-mcp -- docker run --rm -i -e OPA_URL -e OPA_TOKEN -e GITHUB_TOKEN docker.io/orygn/opa-mcp:0.9.1
This image reads OPA_URL, OPA_TOKEN and GITHUB_TOKEN. Set them in your client's env block for this server; docker run -e passes each one through to the container.
What is the OPA MCP server?
OPA MCP is listed in the public MCP registry as io.github.OrygnsCode/opa-mcp. Author, validate, debug, and explain OPA Rego policies through any MCP-compatible client. This page covers its container image (docker.io/orygn/opa-mcp).
Is the OPA MCP server safe to use?
OPA MCP scores 54 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.
What tools does the OPA MCP server expose?
OPA MCP exposes 52 tools: rego_format, rego_check, rego_check_schema, rego_lint, rego_parse_ast, and 47 more. Their descriptions and schemas cost roughly 16,625 tokens of context every time the server is loaded.
Is the OPA MCP server still maintained?
OPA MCP is still listed as active in the MCP registry. We last reached this channel on 5 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.
What licence is the OPA MCP server under?
OPA MCP declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.