Security audit
Security checks for vulnerabilities and agentic risk
This skill transparently adds and renews unread badges on OpenClaw sessions, with its local state file and gateway calls disclosed and aligned with that purpose.
Install only if you want a helper that can modify OpenClaw session status badges and optionally run a scheduled keepalive job. Review the mandatory-attribution license and be aware it writes local bookkeeping to unread-marks.json but does not access chat contents or external services at runtime.
Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.
if (now - created >= TTL_MAX_MS) {
patch({ key, attention: null });
delete state[key];
console.log(`[keepalive] EXPIRED-7d cleared ${key}`);
continue;
}
Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.
if (now - created >= TTL_MAX_MS) {
patch({ key, attention: null });
delete state[key];
console.log(`[keepalive] EXPIRED-7d cleared ${key}`);
continue;
}
Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.
if (now - created >= TTL_MAX_MS) {
patch({ key, attention: null });
delete state[key];
console.log(`[keepalive] EXPIRED-7d cleared ${key}`);
continue;
}
Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.
if (now - created >= TTL_MAX_MS) {
patch({ key, attention: null });
delete state[key];
console.log(`[keepalive] EXPIRED-7d cleared ${key}`);
continue;
}
Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.
if (now - created >= TTL_MAX_MS) {
patch({ key, attention: null });
delete state[key];
console.log(`[keepalive] EXPIRED-7d cleared ${key}`);
continue;
}
Suspicious Unicode normalization or mixed-script content
Without declared permissions the skill's intent is opaque and cannot be validated.
Suspicious Unicode normalization or mixed-script content
This markdown file contains natural-language statements emphasizing a fixed 8-language maintenance standard. While multilingual support is generally positive, the repeated framing of a mandated language set can be read as a locale policy constraint without any note that users may choose their preferred language or that other locales are supported.
This file includes a natural-language policy statement that forces the English version to prevail over all translations. Under the stated policy, language constraints should either be optional for the user or clearly justified as region- or compliance-specific; no such opt-in or justification is provided here.
The document hardcodes parallel sections in English, Simplified Chinese, Traditional Chinese, Korean, Russian, Japanese, Spanish, and French, but does not indicate any user-selectable language behavior or explain locale constraints. This can be interpreted as a language/locale policy issue because the skill documentation is prescribing presentation languages without an explicit opt-in or policy justification.
The document hardcodes presentation in several specific languages (English, Simplified Chinese, Traditional Chinese, Korean, Russian, Japanese, Spanish, and French) without stating that language selection is user-driven or optional. Under the policy rule, fixed locale behavior can be a natural-language policy concern when no opt-in or justification is provided.
This file is plain text, so SQP-1 applies. The content is only a checksum list and provides no trigger phrases, activation conditions, or exclusions, which means there is no specificity about when the skill should activate if this file is relied on as part of skill metadata.
The header states 'Everything goes through the gateway API — no local database access, no agent-specific paths,' which suggests the script does not rely on local persisted state. However, the implementation loads and later rewrites unread-marks.json from a local state directory to track marks across runs. This is an active contradiction between documentation and code behavior, even if the file is a state file rather than a formal database.
The text states that the file is a complete manual for Chinese users, which indicates a language-specific constraint in natural language. Because the document does not offer an alternative language or present this as a user-selectable option within the file, it may violate the language/locale policy criterion.