blaktron/vuln-reachability-audit
autonomousv0.8.0
Audits a codebase's dependencies against OSV.dev, traces reachability in the code, and returns an HTML report with fix suggestions.
Declared capabilities
network: 1 hostdata: code
Capabilities are the author's declaration and the basis for policy and scanning. The registry cannot enforce them at run time. Your runtime should.
MCP servers & tools
agentry run serves exactly the tools this contract declares, through a per-run MCP gateway. An undeclared capability is a tool that does not exist at run time.
Tools the runner grants
| Manifest declaration | Tool(s) granted |
|---|---|
| (always) | write_report — reports → the run's output folder |
| (always) | list_inputs, read_input, search_inputs — operator-provided input files, read-only |
| (always) | scratch_list, scratch_read, scratch_write, scratch_search — the run's own scratch folder at /scratch, private to the run and deleted at its end; the runner's grant, not a line of the contract |
| (always) | export_pdf — a Markdown report as a PDF in reports/, laid out in your report design; the runner's grant, not a line of the contract |
| (always) | grep, glob — search file contents by regular expression and list files by path pattern (ripgrep), one area a call: the inputs, attached knowledge, /scratch; the runner's grant, not a line of the contract |
| network.egress | fetch — pinned to api.osv.dev |
MCP servers involved
- QuickAgent Tools — the runner's companion server, served through a per-run Docker MCP Gateway; its tool surface is exactly the table above.
- Operator-cataloged MCP servers may be attached for the declared dataAccess category (code); undeclared categories are never wired in.
In an embedded install
Installed into an existing coding agent (agentry install), the same contract maps to the host's built-in tools — enforcement there is the host's. Allowed tools: WebFetch(domain:api.osv.dev).
Install
agentry install blaktron/vuln-reachability-audit@0.8.0git clone https://secureagentry.net/blaktron/vuln-reachability-audit.gitgit clone ssh://git@git.secureagentry.net/blaktron/vuln-reachability-audit.gitclaude-skill is this agent's default target, so the flag is not needed.
agent.yaml
2.2 kBBack to files
schemaVersion: "2"
name: vuln-reachability-audit
namespace: blaktron
modes: [autonomous]
version: 0.8.0
summary: Audits repository dependencies against OSV and reports vulnerability reachability, fixes and coverage in HTML.
license: MIT
tags: [vulnerabilities, cve, osv, reachability, dependency-audit, html-report, github]
entrypoints:
instruction: AGENT.md
intake:
- id: repository_url
prompt: Repository HTTPS address, optionally with a branch, tag or full commit; leave blank to audit supplied project files.
type: repository
inputs:
kinds: [code, data]
view: original
guide:
overview: The runner downloads the repository snapshot you select before the audit starts. The agent inventories dependency versions, queries current OSV advisories and traces affected behavior through source and build configuration, returning reports/vuln-reachability.html with findings, remediation and explicit coverage gaps. This is static dependency and reachability analysis, not an executable scanner or a full source-code vulnerability scan; incomplete coverage is prominently labelled.
prepare: ["Provide the repository HTTPS address; include a branch, tag or full commit to select a specific snapshot.", "For private repositories, provide access through the repository source setup; the audit does not receive that credential.", "Include manifests, lockfiles and build configuration for every workspace you want audited.", Optionally supply a current CycloneDX or SPDX JSON SBOM with provenance and deployment/environment assumptions., Large repositories may exceed download or audit limits; completed findings and unfinished scope are reported explicitly.]
files: Optional CycloneDX or SPDX JSON SBOMs and supplemental source/configuration; supplied project files can replace repository intake for code the runner cannot fetch.
task: e.g. Audit all workspaces, including build tooling, and identify which known dependency vulnerabilities are reachable.
runtime:
interfaces: [claude-skill, agents-md]
capabilities:
network:
egress: [api.osv.dev]
filesystem:
read: none
write: none
exec: none
dataAccess: [code]
publish:
include: [agent.yaml, AGENT.md, README.md, LICENSE]