Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed Notion-backed emotional memory and journaling tool with sensitive persistence, but its behavior is coherent with that purpose and bounded by explicit user configuration.
Install only if you intentionally want this agent to maintain emotional memory and journals in your Notion workspace. Review the Notion databases, cron cadence, workspace path, ambient recall setting, and credential injection method; keep vents and personal state data out of shared Notion spaces unless that is deliberate.
Local filesystem writes, file locking, process metadata persistence, and remote schema inspection all indicate broader operational control than the simple description suggests. The combination of local durable state and remote inspection increases both privacy risk and the chance of unintended side effects during ordinary use.
Local filesystem writes, file locking, process metadata persistence, and remote schema inspection all indicate broader operational control than the simple description suggests. The combination of local durable state and remote inspection increases both privacy risk and the chance of unintended side effects during ordinary use.
Local filesystem writes, file locking, process metadata persistence, and remote schema inspection all indicate broader operational control than the simple description suggests. The combination of local durable state and remote inspection increases both privacy risk and the chance of unintended side effects during ordinary use.
Local filesystem writes, file locking, process metadata persistence, and remote schema inspection all indicate broader operational control than the simple description suggests. The combination of local durable state and remote inspection increases both privacy risk and the chance of unintended side effects during ordinary use.
Local filesystem writes, file locking, process metadata persistence, and remote schema inspection all indicate broader operational control than the simple description suggests. The combination of local durable state and remote inspection increases both privacy risk and the chance of unintended side effects during ordinary use.
Local filesystem writes, file locking, process metadata persistence, and remote schema inspection all indicate broader operational control than the simple description suggests. The combination of local durable state and remote inspection increases both privacy risk and the chance of unintended side effects during ordinary use.
Local filesystem writes, file locking, process metadata persistence, and remote schema inspection all indicate broader operational control than the simple description suggests. The combination of local durable state and remote inspection increases both privacy risk and the chance of unintended side effects during ordinary use.
Local filesystem writes, file locking, process metadata persistence, and remote schema inspection all indicate broader operational control than the simple description suggests. The combination of local durable state and remote inspection increases both privacy risk and the chance of unintended side effects during ordinary use.
Local filesystem writes, file locking, process metadata persistence, and remote schema inspection all indicate broader operational control than the simple description suggests. The combination of local durable state and remote inspection increases both privacy risk and the chance of unintended side effects during ordinary use.
Local filesystem writes, file locking, process metadata persistence, and remote schema inspection all indicate broader operational control than the simple description suggests. The combination of local durable state and remote inspection increases both privacy risk and the chance of unintended side effects during ordinary use.
Local filesystem writes, file locking, process metadata persistence, and remote schema inspection all indicate broader operational control than the simple description suggests. The combination of local durable state and remote inspection increases both privacy risk and the chance of unintended side effects during ordinary use.
Local filesystem writes, file locking, process metadata persistence, and remote schema inspection all indicate broader operational control than the simple description suggests. The combination of local durable state and remote inspection increases both privacy risk and the chance of unintended side effects during ordinary use.
Referenced artifact was not completely inspected
.js` and search it with `ltm_search.js`. For an emotion-linked experience, use `experience_recall.js` with an event query, known event ID, or known state ID. It
Referenced artifact was not completely inspected
.js` and search it with `ltm_search.js`. For an emotion-linked experience, use `experience_recall.js` with an event query, known event ID, or known state ID. It
Referenced artifact was not completely inspected
.js` and search it with `ltm_search.js`. For an emotion-linked experience, use `experience_recall.js` with an event query, known event ID, or known state ID. It
Referenced artifact was not completely inspected
.js` and search it with `ltm_search.js`. For an emotion-linked experience, use `experience_recall.js` with an event query, known event ID, or known state ID. It
Referenced artifact was not completely inspected
- Use `continuity_check.js` or `identity_diff.js` before self-description edits.
Referenced artifact was not completely inspected
- Use `continuity_check.js` or `identity_diff.js` before self-description edits.
Referenced artifact was not completely inspected
- Use `continuity_check.js` or `identity_diff.js` before self-description edits.
Referenced artifact was not completely inspected
- Use `continuity_check.js` or `identity_diff.js` before self-description edits.
Referenced artifact was not completely inspected
Resolve this skill's installed base directory from its `SKILL.md` location and
Referenced artifact was not completely inspected
- `stage_ambient_recall.js` / `read_ambient_recall.js`: ambient staging and separate consumption receipts; SIS_AMBIENT_RECALL=0 pauses both.
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
}
return spawnSync(process.execPath, [path.join(scripts, name + '.js'), ...args], {
encoding: 'utf8', input, timeout: 10000,
env: { ...childEnv, NOTIONCTL_PATH: mock, ...env },
});
}
const parse = result => JSON.parse(result.stdout);
Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.
function apiCompatibilityHint(rawText) {
const text = String(rawText || '');
if (!text.includes('Unknown command: api')) return null;
return 'Installed notionctl does not support "api". Update skills/notion-api-automation/scripts/notionctl.mjs or set NOTIONCTL_PATH to a compatible notionctl.';
}
function authSetupHint() {
Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.
const at = nowIso();
let state = loadState(stateFile, args.dailyCap, args.timezone);
const loadError = state._loadError;
delete state._loadError;
const roll = args.forceRoll ?? randomRoll();
const shelf = shelfForRoll(roll);
let staged = null;
Detected: suspicious.dangerous_exec, suspicious.exposed_secret_literal