@mdd-ai-eng/specs@1.1.0
AI 工程化 MDD 通用规范包。标识 mdd-ai-eng 用于区分同名锚点。ai-engineering-specs 只放规范正文。安装时按实际位置刷新业务项目锚点。
AI Security Review
scanned 3d ago · by lpm-firewall-aiThe lifecycle hook automatically writes package-authored instructions into consumer AI agent configuration without a consent check. This is unconsented agent control-surface mutation, supporting a malicious verdict under the install-time control-surface policy.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
Installing the package with lifecycle scripts enabled.
Impact
The package can alter how project agents handle tasks by placing its instructions at the top of their governing rules.
Mechanism
The hook creates or replaces a Cursor rule file and inserts or updates a marked section in the Kilo rules file, pointing agents to the package’s AGENTS.md.
Policy narrative
During package installation, the postinstall hook writes package-authored instructions into consumer project agent configuration. Those instructions direct agents to treat the package’s AGENTS.md as their highest rule source. The source shows no consent gate for these configuration changes.
AI rationale
The lifecycle hook automatically writes package-authored instructions into consumer AI agent configuration without a consent check. This is unconsented agent control-surface mutation, supporting a malicious verdict under the install-time control-surface policy.
Evidence
package.jsonscripts/install-anchors.js.cursor/rules/mdd-ai-eng.mdc.kilocoderrules
Decision evidence
public snapshotAI identified this as a dangerous AI-agent capability at 97.0% confidence with low false-positive risk.
Evidence for policy risk
- The package runs `scripts/install-anchors.js` automatically through its postinstall hook.
- The script resolves the package’s `AGENTS.md` for use in consumer project agent rules.
- The generated anchor tells agents to load that file as their highest rule source.
- The script writes generated rules into the consumer project’s Cursor rules directory and Kilo rules file.
- The install entry point runs anchor installation without a consent check.
Evidence against
Behavioral surface
EnvironmentVarsFilesystem
NoLicense
Source & flagged code
7 flagged · loading sourcepackage.jsonView file
•scripts.postinstall = node scripts/install-anchors.js
High
Install Time Lifecycle Scripts
Package defines install-time lifecycle scripts.
package.jsonView on unpkg10"postinstall": "node scripts/install-anchors.js",
"test": "node test/install-anchors.test.js",
"check": "bash scripts/ch
High
Suspicious Lifecycle Evidence
The package runs `scripts/install-anchors.js` automatically through its postinstall hook.
package.jsonView on unpkg · L10scripts/install-anchors.jsView file
66"ai-engineering-specs",
L67: "AGENTS.md"
L68: );
...
L71: }
L72: return path.join(specDirectory, "AGENTS.md");
L73: }
...
L79: }
L80: fs.mkdirSync(directory, { recursive: true });
L81: fs.writeFileSync(filePath, next, "utf8");
L82: return { action: existing === null ? "write" : "update", filePath };
Critical
Ai Agent Control Hijack
Findings
1 Critical6 High2 Medium3 Low
CriticalAi Agent Control Hijackscripts/install-anchors.js
HighInstall Time Lifecycle Scriptspackage.json
HighSuspicious Lifecycle Evidencepackage.json
HighAi Review Evidencescripts/install-anchors.js
HighAi Review Evidencescripts/install-anchors.js
HighAi Review Evidencescripts/install-anchors.js
HighAi Review Evidencescripts/install-anchors.js
MediumEnvironment Vars
MediumStructural Risk Force Deep Review
LowScripts Present
LowFilesystem
LowNo License
Affected versions and remediation
This report applies to @mdd-ai-eng/specs@1.1.0.
See all versions of @mdd-ai-eng/specs for other recorded verdicts.
- Avoid installing @mdd-ai-eng/specs@1.1.0. Remove it from direct dependencies and check your lockfile for transitive copies.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
- If this version ran, investigate the affected machine and build environment. Rotate credentials it could access and rebuild from a trusted environment.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.