[sighthound] Security findings in github/gh-aw

[github-actions]

Sighthound scan found 148 findings (116 outside test paths). Top 5 by severity below.

1. Command Injection — pkg/cli/runner_guard.go:114

• Severity: Critical (High confidence)
• exec.Command(dockerPath, "run", ..., "-v", volumeMount, ..., containerScanPath, ...) uses exec.Command with a fixed slice, but volumeMount/containerScanPath are built from dynamic paths.
• Remediation: validate/sanitize volumeMount and containerScanPath inputs before use; ensure they cannot contain shell metacharacters or unexpected path traversal, and confirm dockerPath is resolved from a trusted location.

2. Command Injection — pkg/cli/grant.go:176

• Severity: Critical (High confidence)
• Similar exec.Command(dockerPath, "run", ..., volumeMount, ..., containerPolicyPath, ..., imageRef) pattern with dynamic imageRef/paths.
• Remediation: validate imageRef and mount paths against an allowlist/pattern before passing to the command.

3. Command Injection — pkg/cli/poutine.go:229

• Severity: Critical (High confidence)
• exec.Command(dockerPath, "run", ..., volumeMount, ...) with dynamic volume mount value.
• Remediation: sanitize/validate volumeMount path construction; avoid interpolating unsanitized user/repo paths.

4. Command Injection — pkg/cli/poutine.go:109

• Severity: Critical (High confidence)
• Related invocation building docker command args from dynamic data.
• Remediation: same as above — validate all dynamic arguments passed into exec.Command.

5. Command Injection — pkg/cli/upgrade_command.go:507

• Severity: Critical (High confidence)
• Command arguments built from untrusted/dynamic data (e.g., version/upgrade parameters).
• Remediation: validate version strings and other dynamic inputs against strict patterns (e.g., semver regex) before passing to exec.Command.

---

Note: these all use Go's exec.Command with argument slices (not sh -c), which mitigates typical shell injection, but the flagged variables (paths, image refs, versions) should still be validated since they may originate from untrusted sources (e.g., workflow inputs, external configs).

Findings limited to non-test files from actionable.json; full scan produced 148 findings total.

Generated by 🛡️ Sighthound Security Scan · auto · 19 AIC · ⌖ 2.61 AIC · ⊞ 7.6K · ◷

[github-actions]

🍪 Issue Monster selected this for Copilot\n\nI've identified this issue as a good candidate for automated resolution and requested assignment to the Copilot coding agent.\n\nIf assignment succeeds, the Copilot coding agent will analyze the issue and create a pull request with the fix.\n\nOm nom nom! 🍪

🍪 Om nom nom by Issue Monster · gpt54 · 2.73 AIC · ⌖ 8.54 AIC · ⊞ 10.1K · ◷