Productivity

• Outcome: Review
• Latest audit: Jul 27, 2026 · 10:16 AM UTC
• Version: 1.0.6

Overview

This productivity skill is coherent and local-only, but it automatically stores long-lived personal, work, contact, and health notes without a clear consent step.

Install only if you want an agent to maintain a persistent local productivity memory for you. Review or back up ~/Clawic/data/ first, and be especially careful before letting it store health facts, medication timing, client/customer details, or personal behavior patterns. It does not show evidence of network exfiltration or credential storage, but it is designed to write durable records automatically.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (12)

Context-Inappropriate Capability

Medium

• Confidence: 93% confidence
• Finding: The skill explicitly instructs persisting user health and productivity details across multiple files, including shared and long-lived storage, which goes beyond transient advisory use. This creates unnecessary retention of sensitive personal data and broadens exposure if other skills, users, or processes can access those files later.

Description-Behavior Mismatch

Medium

• Confidence: 91% confidence
• Finding: The skill instructs writes to multiple files outside the core productivity memory, including shared config and cross-skill data stores such as contacts and artifacts. Even though the content is productivity-related, this expands the skill’s authority boundary and can cause unintended persistence, cross-skill state corruption, or leakage of user/workplace metadata into locations other skills may consume.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The skill explicitly instructs storing medication timing or an ADHD diagnosis in a shared health profile, which is sensitive health information. Although it limits recording to user-stated facts and avoids diagnosis, it does not require an explicit privacy warning, consent check, data-minimization guidance, or retention/access controls before persisting that information.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The file explicitly instructs recording diagnosis, medication, or clinician-set restrictions in a shared health profile, which is sensitive health data. Even though it says not to infer symptoms, it lacks an explicit privacy warning, consent requirement, access-control guidance, or data-minimization limits, creating a real risk of unnecessary collection and broader exposure of regulated or highly sensitive personal information.

Missing User Warnings

Medium

• Confidence: 97% confidence
• Finding: This line directs the agent to write diagnoses, medications, sleep disorders, and other health facts into a shared profile file without any explicit warning, consent checkpoint, or clear access boundary in this skill. Health information is highly sensitive, and placing it in shared memory increases the risk of secondary use, unintended disclosure, and cross-context leakage to unrelated workflows.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The skill explicitly instructs the agent to read `~/Clawic/data/productivity/memory.md` and the shared `~/Clawic/data/projects/` before advising, but it provides no privacy notice, data-minimization rule, or user-consent check. This is dangerous because those files may contain sensitive personal or business information, and the skill normalizes broad file access beyond what is strictly necessary for the immediate request.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The skill explicitly instructs writing to and creating files under the user's data directory, including sessions logs, config.yaml, and artifact files, without any user-facing confirmation or warning at the point of action. Even though the purpose is productivity tracking, silent persistence can overwrite user data, create unwanted records, or mutate configuration in ways the user did not knowingly approve.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The template explicitly authorizes writes and deletions without requiring user confirmation, which can cause unintended modification or loss of user data if the agent misclassifies information or acts on ambiguous input. In a memory-management skill that continuously rewrites files, silent mutation is more dangerous because the user may not realize durable state has changed until later.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The split procedure instructs the agent to delete sections from memory.md in the same turn as moving content, with no rollback or verification step. If the move is incomplete, malformed, or targets the wrong section, this can permanently erase user memory data and leave inconsistent indexes.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The skill explicitly instructs the agent to read multiple local memory and project files, including personal productivity memory that may contain sensitive life details such as therapy, school runs, care schedules, and commitments, without any user-facing consent prompt or data-minimization step. In this context, the data access is functionally relevant to planning, but it still creates an unnecessary privacy risk because the skill normalizes broad file access before establishing whether all referenced content is needed for the current request.

Ssd 3

Medium

• Confidence: 93% confidence
• Finding: The file explicitly instructs the agent to persist user-specific behavioral preferences and recurring emotional patterns into shared memory files. Even though it forbids recording clinical information, these notes still constitute sensitive profiling data that can reveal mental state, work struggles, and personal vulnerabilities, and storing them in shared/shared-across-sessions memory increases privacy and misuse risk.

Autonomous Decision Making

Medium

• Category: Excessive Agency

Content

**Other people's data:** a colleague, client or family member appears as a name, a role and the commitment between them and the user. Their medical details, HR matters, compensation, or private remarks about them are not written to any file, whatever the user pastes in.

**Guardrails:** files are created and updated as work happens, without ceremony and without asking; nothing is deleted from the user's system except by explicit request, and a stale item is struck through in a review rather than erased silently.

## Related Skills
More Clawic skills, get them at https://clawic.com/skills/productivity (install if the user confirms):

• Confidence: 96% confidence
• Finding: without asking

VirusTotal

VirusTotal findings are pending for this skill version.

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical

• Code: suspicious.exposed_secret_literal
• Location: memory-template.md:67