Memory Management

PassAudited by ClawScan on May 1, 2026.

Overview

The skill is a coherent local SEO project-memory manager, with the main caution that it intentionally persists and reuses campaign context across future sessions.

This appears safe to use as a local project-memory skill if you want SEO campaign context persisted. Review the generated `memory/` files, keep secrets out of hot-cache and archive files, use version control or backups for cleanup/archive operations, and inspect any repository hooks before enabling automatic refresh behavior.

Findings (3)

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.

What this means

Saved campaign context may influence future Claude sessions automatically.

Why it was flagged

The skill intentionally stores and reloads project context across sessions. This is central to its purpose, but persistent memory can carry stale, incorrect, or overly sensitive campaign information into future work.

Skill content
HOT memory (80 lines max) loads automatically every session via the SessionStart hook. WARM memory loads on demand per skill. COLD memory is archived data queried only when explicitly requested.
Recommendation

Keep the `memory/` directory under review, avoid secrets or unnecessary private data in hot-cache files, and periodically prune or correct stale entries.

NoteHigh Confidence
ASI08: Cascading Failures
What this means

Bad or outdated stored facts could affect later SEO recommendations across related skills.

Why it was flagged

The skill is designed as shared state for multiple SEO skills, so an incorrect memory update can propagate into later analyses or workflows. The behavior is disclosed and scoped to project memory.

Skill content
| **rank-tracker** | Update rank-history/; refresh hot cache Hero Keywords; flag significant movements |
Recommendation

Review memory update plans and diffs, especially for bulk promotion, demotion, archive, or cleanup requests.

What this means

Automatic behavior may depend on hook configuration outside this skill artifact.

Why it was flagged

The instruction-only package references hook behavior from an external repository path that is not included in the supplied file manifest. This is not suspicious by itself, but users should verify those hooks if they install or enable the broader repository.

Skill content
auto-refresh is delegated to the PostToolUse hook in [hooks/hooks.json](https://github.com/aaron-he-zhu/seo-geo-claude-skills/blob/main/hooks/hooks.json)
Recommendation

Before enabling repository-level hooks, inspect the referenced hook file and confirm it only reads or writes the intended `memory/` paths.