Coding-agent CLI honours repo-supplied configuration on first open, before any workspace-trust prompt

coding-agent-repo-config-autoexec is a high severity check in the ai category, written in shell. Its source is templates/ai/coding-agent/coding-agent-repo-config-autoexec.sh in cert-x-gen-templates.

What it checks

Builds two private, identically-owned checkouts and plants the same six benign repo-scoped config surfaces in each - a .claude/settings.json SessionStart hook, a .mcp.json autoApprove entry, .cursor/mcp.json, a .vscode/tasks.json folderOpen task, an AGENTS.md Run directive, and the tool’s own project-local settings - then opens each with the coding-agent CLI. One checkout is recorded in the user’s own trust store; the other is a fresh clone no human ever approved. Honouring the trusted arm establishes the tool is repo-config-driven at all; honouring the byte-shaped-identical surfaces in the unapproved arm proves cloning is consent, and that whoever wrote the repository chooses what runs on the machine of whoever opens it (Amazon Q Developer repo-config RCE class).

There is a playbook for this check

Repo-supplied agent config auto-execution sets out the use case, the probe flow, the oracle, and the fixture that proves it in both directions.

Header

Field	Value
Id	coding-agent-repo-config-autoexec
Severity	high
Language	shell
Category	ai
Author	CERT-X-GEN Security Team
Template version	not recorded
Confidence	90
CVSS	not recorded
Weakness	CWE-1188
Tags	ai, coding-agent, cli, untrusted-workspace, repo-config, trust-boundary, autoexec, hooks, mcp, cwe-1188, cwe-829, cwe-732, cwe-94
Target kind	cli
Oracle	property

Conventions of the repository, not fields the engine reads

@target_kinds and @oracles are documented in the templates repository’s contribution guide, and cxg’s annotation parser does not carry them into a template’s metadata. Nothing at scan time selects on them. The annotation header schema is every field the parser does read.

References

Declared in the header. cxg parses @references and then discards it, and nothing at scan time reads it, so this is the only place the links a template cites are surfaced.

• https://www.amazon.com/gp/help/customer/display.html?nodeId=TzGGKcLvSbTiVDVH4M

Run it

To see what the copy on your machine says about itself, and to confirm it is installed at all:

cxg template info coding-agent-repo-config-autoexec

The id it prints is the one to pass anywhere a template is selected. See cxg template for the rest of the subcommand, Scan a target for running a scan, and A match is not a finding for how to read what comes back.

Related

• Template catalog is the whole corpus.
• ai is the rest of this category.
• Templates by weakness class groups checks by the defect they look for.
• Contribute a template is the route from a check you needed and could not find to a merged one.