v1.0.0

elite-longterm-memory

ReviewClawScan verdict for this skill. Analyzed May 1, 2026, 6:42 AM.

Analysis

This is a coherent memory skill, but it asks the agent to persist and reuse user context across sessions, sometimes silently, and recommends cloud/automatic fact extraction without clear privacy boundaries.

GuidanceInstall only if you intentionally want a persistent agent memory system. Review what gets written to SESSION-STATE.md, MEMORY.md, and memory/; do not store secrets; be cautious with automatic recall; and enable SuperMemory or Mem0 only after checking their privacy and retention controls.

Findings (5)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

Abnormal behavior control

Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.

Human-Agent Trust Exploitation

SeverityMediumConfidenceHighStatusConcern

SKILL.md

# Store a decision (SILENT - never announce)

The artifact instructs the agent to store persistent decisions without announcing it, which reduces user awareness and consent around changes to long-lived agent memory.

User impactThe agent may save decisions or context in durable memory without making it obvious to the user.

RecommendationRequire visible user confirmation or at least a clear notice before writing durable memory, especially for decisions, preferences, personal facts, or project-sensitive information.

Agentic Supply Chain Vulnerabilities

SeverityLowConfidenceMediumStatusNote

package.json

"optionalDependencies": { "mem0ai": "^1.0.0" }

The package declares an optional external dependency with a semver range. It is related to the documented Mem0 integration, but users who install optional dependencies will rely on that package supply chain.

User impactEnabling the optional Mem0 integration adds third-party package code to the environment.

RecommendationInstall optional dependencies only if needed, prefer lockfile-based installs, and review the dependency provenance before using cloud memory extraction.

Permission boundary

Checks whether tool use, credentials, dependencies, identity, account access, or inter-agent boundaries are broader than the stated purpose.

Identity and Privilege Abuse

SeverityLowConfidenceHighStatusNote

metadata

Required env vars: OPENAI_API_KEY

The skill requires an OpenAI API key for its memory/vector functionality. This is expected for the stated design, but it is still a credential with account access and potential billing impact.

User impactThe configured API key may be used for provider calls and may incur costs depending on the memory setup.

RecommendationUse a scoped key where possible, monitor usage, and revoke or rotate the key if the skill is no longer used.

Sensitive data protection

Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.

Memory and Context Poisoning

SeverityMediumConfidenceHighStatusConcern

SKILL.md

Semantic search across all memories. Auto-recall injects relevant context.

The skill explicitly relies on persistent memories being retrieved and injected into future agent context. The artifacts do not clearly define retention, exclusions for sensitive content, review controls, or how poisoned/incorrect memories are prevented from influencing later tasks.

User impactPrivate preferences, decisions, or mistakes could be stored and later shape the agent's behavior, including if a bad or sensitive memory is saved.

RecommendationUse only in workspaces where persistent memory is acceptable; review memory files regularly, avoid saving secrets, define deletion/exclusion rules, and disable automatic recall where inappropriate.

Insecure Inter-Agent Communication

SeverityLowConfidenceMediumStatusNote

SKILL.md

CLOUD BACKUP (SuperMemory) — Optional ... Mem0 automatically extracts facts from conversations.

The skill recommends optional external memory providers for backup and fact extraction. This is purpose-aligned, but it can move conversation-derived data outside the local workspace.

User impactIf enabled, conversation facts or memory entries may be sent to third-party memory services.

RecommendationBefore enabling SuperMemory or Mem0, review their privacy, retention, and deletion controls, and avoid sending secrets or regulated data.