Agent Receipt Validator Mcp — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited Agent Receipt Validator Mcp (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
A local stdio MCP server for validating signed agent execution receipts against evidence bundles.
This server wraps the verifiable-tool-invocation-flow Python package and exposes receipt validation as MCP tools for local clients such as Claude Desktop, Cursor, and other MCP-compatible agent runtimes.
execution_receipt.json payload against an evidence_bundle.json payload and a public key PEM.pip install agent-receipt-validator-mcp==0.1.1cd agent-receipt-validator-mcp
python3.13 -m venv .venv
. .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install -e ".[test]"Use this MCP client stdio configuration after installing the package in your local environment:
{
"mcpServers": {
"agent-receipt-validator": {
"command": "agent-receipt-validator-mcp",
"args": []
}
}
}validate_receipt_jsonInputs:
receipt_json: JSON string containing an execution receipt.evidence_bundle_json: JSON string containing the evidence bundle.public_key_pem: public key PEM string.audience: expected audience, default demo-validator.Output:
{
"verdict": "valid",
"report": {
"verdict": "valid"
}
}generate_demo_artifactsInputs:
audience: expected audience, default demo-validator.Output:
{
"receipt_json": "{...}",
"evidence_bundle_json": "{...}",
"public_key_pem": "-----BEGIN PUBLIC KEY-----...",
"verification_report_json": "{...}",
"verdict": "valid"
}The demo artifacts are generated fresh at runtime. The server does not use static expired samples.
summarize_verification_reportInputs:
verification_report_json: JSON string containing a verification report.Output:
{
"verdict": "invalid",
"failed_checks": ["tool_output_hash_match"],
"warnings": [],
"errors": ["tool_output_hash_mismatch"]
}This MCP server validates signed execution evidence. It does not prove semantic correctness of the tool output. It does not prove that the policy itself is correct. It does not protect against a compromised signer. It does not replace sandboxing, IAM, access control, monitoring, or human approval. Do not pass private keys, API tokens, confidential evidence bundles, or production receipts to untrusted MCP clients.
This first release is a local stdio MCP server. Remote Streamable HTTP hosting and Smithery publication are future tasks.
This package is intended to be published through PyPI Trusted Publishing using GitHub Actions OIDC.
Workflow:
.github/workflows/publish.ymltestpypipypiNo PyPI API token is stored in this repository.
Trusted Publishing must be configured once in TestPyPI/PyPI before the workflow can publish.
pytest tests/test_tools.py
agent-receipt-validator-mcp --help~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.