
MCP Blast-Radius Auditor
io.github.aos-standardPublished by GitHub account aos-standardv0.2.8Updated Oct 24, 2026
Static blast-radius audit for MCP servers before agent install. Manifest optional.
The publisher's own server: Blast Radius
Overview
Statically audits an MCP server's source to show what it can reach — files, network, subprocess, env — and flags divergences from its declared manifest.
- What you can do
- Scan one MCP server directory and return a blast-radius report of filesystem, network, subprocess and environment access
- Compare observed access against a declared manifest and list divergences in blocking reasons
- Run a wiring smoke test of the auditor itself
- Report dependency declarations from requirements, pyproject and package.json files
- Label each finding as declared, observed-static or cannot-determine
- When to use it
- Use it before adding a third-party MCP server to an agent, or in CI to block merges when a server touches paths or capabilities it did not declare. Useful when you want a static, upper-bound view of a package's reach without running it.
- Requirements
- Runs as a local Python package (installable with pip or uvx) over stdio, for desktop clients. You supply the target directory to scan, either as the target_dir argument or via the AOS_VALIDATOR_TARGET_DIR environment variable. Optional variables AOS_VALIDATOR_MCP_LOG and AOS_VALIDATOR_CALLER set a local call log path and caller label.
Installation
In SourceWeft
- Open MCP Blast-Radius Auditor in the dashboard and add it to a workspace.
- Enable the server for the chats that should use its tools.
Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.
Other MCP clients
Follow the launch instructions in the repository.
README
MCP Blast-Radius Auditor
See what any MCP server can actually touch — before you add it to your agent.
No manifest? You still get the full blast-radius report. Add a manifest to also catch divergences.
Also, if the server declares a manifest: Catch an MCP server that touches files it said it wouldn't — and block the merge in CI.
Statically extract what a third-party MCP server can reach (files, network, subprocess, env) via surface-level analysis. Compare against declared boundaries when a manifest is present.
Scan scope (default): production package only — excludes tests/, docs/, examples/, scripts/, benchmarks/, .github/, and test_*.py patterns; JSON output includes scan_scope and excluded_file_count. Pass --include-peripheral to scan the full repo.
Try it in 3 steps
① Scan your server in one command
Point --target-dir at your shipping package root (e.g. src/). Default scope excludes tests, docs, and scripts.
② Read the JSON
Undeclared capability is usually drift, not malice. Treat network/subprocess counts as upper bounds, not confirmed traffic.
③ Apply for an audit badge (optional, opt-in)
Ran a clean scan and want a signed README badge? Open a badge application — paste your command and JSON. Free, 90-day attestation, no phone-home. Criteria: BADGE_CRITERIA.md.
To verify any published attestation independently: pip install cryptography, then run packaging/scripts/verify_attestation.py (accepts local paths or HTTPS URLs). See BADGE_CRITERIA.md §Verify.
Machine-readable metadata
- Agent Card (capabilities, limitations, pricing): agent_card.json
- Catalog entry (pricing, install, MCP endpoint): aos-standard/catalog
- Spec: AOS-v0.1
Example walkthrough
Inspect blast_radius and any DIVERGENCE: lines in blocking_reasons.
Report a scan question
Open a GitHub issue with your JSON output (structured template loads automatically).
30-second scan
pipx run mcp-blast-radius starts the MCP stdio server (for Claude Desktop / Cursor). For CLI scanning, use mcp-blast-radius-gate as above.
- Red (blocking): divergence detected — code touches paths or capabilities not declared in manifest.
- Green: no divergences (or no manifest — blast radius report only, advisory pass).
Install
CLI entry
CI blocking gate
MCP tools
aos_compliance_validate— scan one MCP server directory (target_dirrequired;tool_idoptional label)aos_compliance_self_test— wiring smoke test
Default gate_mode=advisory. Use gate_mode=blocking in CI to fail on divergences.
What is extracted
Limitations: Static analysis only. Dynamic imports, getattr/eval, obfuscation, and native extensions may hide capabilities. We do not claim complete coverage — every finding includes a confidence label.
Environment
Example
License
MIT
Source: packaging/README.md at commit d75ea1c
Tools
0Version history
2- v0.2.8LatestSep 23, 2026
- v0.2.4Sep 16, 2026
Score
50% of the applicable checks met
How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.
- Reached in the last 30 days25 pts
Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.
- Auth declared10 pts
Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.
- Tools listed15 pts
Answer tools/list without a credential, or ship a package that starts in a sandbox without one.
- DescriptionRequired15 pts
- Logo10 pts
- Documentation15 pts
- License10 pts
Other Blast-Radius integrations

Blast Radius
ai.blast-radiusOfficial publisher
Lets an assistant read Blast Radius weekly AI market arguments, headlines, and an ETF quote, and assess companies with its framework.

Agent Blast Radius
Published by GitHub account makash
Scan what an AI agent running as you can reach; optionally verify which keys are live (x402).