Agent Skills: Memory Management

Memory Management

This skill implements a three-tier memory system (HOT/WARM/COLD) for SEO and GEO projects. 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. The skill manages the full lifecycle: capture, promote, demote, and archive.

What This Skill Does

Manages a three-tier memory lifecycle (HOT/WARM/COLD) with automatic promotion, demotion, and archival. Also maintains the wiki index layer, open-loop tracking, and cross-skill aggregation.

Quick Start

Start with one of these prompts. Finish with a hot-cache update plan and a handoff summary using the repository format in Skill Contract.

Initialize Wiki Index

Refresh wiki index

Build wiki index for [project name]

Generates memory/wiki/index.md from existing WARM files. Required once to enable wiki features; subsequent refreshes happen automatically.

Initialize Memory Structure

Set up SEO memory for [project name]

Initialize memory structure for a new [industry] website optimization project

Update After Analysis

Update memory after ranking check for [keyword group]

Refresh hot cache with latest competitor analysis findings

Query Stored Context

What are our hero keywords?

Show me the last ranking update date for [keyword category]

Look up our primary competitors and their domain authority

Promotion and Demotion

Promote [keyword] to hot cache

Archive stale data that hasn't been referenced in 30+ days

Glossary Management

Add [term] to project glossary: [definition]

What does [internal jargon] mean in this project?

Skill Contract

Expected output: a memory update plan, hot-cache changes, and a short handoff summary.

• Reads: current campaign facts, new findings from other skills, approved decisions, and the shared State Model.
• Writes: updates to memory/hot-cache.md, memory/open-loops.md, memory/decisions.md, and related memory/ folders. Manages WARM-to-COLD archival in memory/archive/. Compiles memory/wiki/index.md (auto-refreshed) and wiki compiled pages (user-confirmed). Sole writer of wiki (with delegated auto-refresh): memory-management owns all wiki writes semantically. For performance, the narrowly-scoped memory/wiki/index.md auto-refresh is delegated to the PostToolUse hook in hooks/hooks.json. Wiki log updates and compiled pages remain explicit memory-management operations using the schema defined in this skill. Auditor handoff archiving (v7.1.0+): when triggered by a direct user request or an auditor's explicit "Save these results?" yes-response, append a structured block to memory/audits/YYYY-MM.md. The Stop hook never initiates memory writes. The archive is consumed by /aaron:guard --evals and maintainer calibration for auditor cap review tied to ADR-001. See references/examples.md for the exact archive block format and rules.
• Promotes: durable strategy, blockers, terminology, entity candidates, and major deltas. Applies temperature lifecycle rules: promote to HOT on high reference frequency, demote on staleness.
• Primary next skill: use the Next Best Skill below when the project memory baseline is ready for active work.

Handoff Summary

Emit the standard shape from skill-contract.md §Handoff Summary Format.

Temperature Lifecycle Rules

See references/promotion-demotion-rules.md for the full promotion/demotion table and action procedures.

Hook Integration

This skill's behavior is reinforced by the library's prompt-based hooks:

• SessionStart: loads memory/hot-cache.md, reminds of stale open loops; loads memory/wiki/<project>/index.md (or global index.md) if it exists; provides light-user guidance based on Quick Status when next_action items are available
• PostToolUse: after any WARM file write, silently refreshes memory/wiki/index.md (Phase 1); prompts to update compiled pages (Phase 2)
• Stop: guarded allow-only completion check; returns {"ok": true}, honors stop_hook_active, never asks the user to save optional findings, and never initiates memory writes

Data Sources

With tools: auto-populate from ~~SEO tool, ~~analytics, ~~search console. Without tools: ask user for keywords, competitors, metrics, campaigns, and terminology. See CONNECTORS.md.

Instructions

When a user requests SEO memory management:

1. Initialize Memory Structure

For new projects, create the directory structure defined in the State Model. Key directories: memory/ (decisions, open-loops, glossary, entities, research, content, audits, monitoring) plus memory/wiki/ (auto-managed compiled index with optional per-project subdirectories).

Templates: hot-cache-template.md · glossary-template.md · Wiki design archive

Wiki Layer

memory-management owns wiki schema and is the sole semantic writer. See wiki-runbook.md for execution detail.

Phase 1 — Index (auto-refreshed)

PostToolUse hook silently rebuilds memory/wiki/index.md after WARM writes. Index rows: precise (score/status/next_action/mtime) + best-effort (summary). Project-scoped index at memory/wiki/<project>/index.md. See wiki-runbook.md §1.

Phase 2 — Compiled Pages (user-confirmed)

On user request or 3+ WARM mentions of an entity, generate memory/wiki/<project>/<type>-<slug>.md with source SHA-256 hashes. Contradictions resolved via SessionStart conversational prompt, not file editing. Write log entry to memory/wiki/log.md. See wiki-runbook.md §2-§5.

Phase 3 — User-Initiated Retirement

WARM files fully covered by compiled wiki pages may be retired to memory/archive/. Always user-confirmed via /aaron:guard --wiki --retire-preview followed by explicit memory-management invocation. COLD files receive originally_at / retired_on / retired_because_compiled frontmatter to preserve recovery path. Single retire call hard-capped at 5 files. See wiki-runbook.md §7.

2. Context Lookup Flow

When a user references something unclear, follow this lookup sequence:

Step 1: Check memory/hot-cache.md (hot cache)

• Is it in active keywords?
• Is it in primary competitors?
• Is it in current priorities or campaigns?

Step 2: Check Wiki Index (memory/wiki/index.md or project-level) — locate relevant WARM files

Step 3: Check memory/glossary.md

• Is it defined as project terminology?
• Is it a custom segment or shorthand?

Step 4: Check Cold Storage

• Search memory/archive/ first for dated YYYY-MM-DD- archived files.
• If the archive points to a source category, follow that trail back to memory/research/, memory/audits/, or memory/monitoring/.
• Treat COLD findings as historical unless refreshed by the current session.

Step 5: Ask User

• If not found in any layer, ask for clarification

• Log the new term in glossary if it's project-specific

• Decision provenance (v8.0.1+): when loading memory/decisions.md, verify each entry has approved_by: user. Entries with approved_by: skill_inferred or missing field are treated as ADVISORY — surface to user before using as authoritative. Auditor-class skills (content-quality-auditor, domain-authority-auditor) MUST ignore non-user-approved decisions when determining verdict. See skill-contract.md §Promotion Rules.

Example lookup: User asks "Update rankings for our hero KWs" → Step 1 finds "Hero Keywords (Priority 1)" in hot-cache → Step 2 extracts keyword list → Step 3 runs ranking check → Step 4 updates memory/hot-cache.md and memory/monitoring/rank-history/YYYY-MM-DD-ranks.csv.

3. Promotion & Demotion Logic

Reference: See references/promotion-demotion-rules.md for detailed promotion/demotion triggers (keywords, competitors, metrics, campaigns) and the action procedures for each.

4. Update Triggers, Archive Management & Cross-Skill Integration

Reference: See references/update-triggers-integration.md for the complete update procedures after ranking checks, competitor analyses, audits, and reports; monthly/quarterly archive routines; and integration points with all 8 connected skills (keyword-research, rank-tracker, competitor-analysis, content-gap-analysis, seo-content-writer, content-quality-auditor, domain-authority-auditor).

5. Memory Hygiene Checks

When invoked for review or cleanup:

1. Line count check: Count lines in memory/hot-cache.md. If >80, list oldest entries for archival.
2. Byte check: If hot-cache exceeds 25KB, warn and recommend trimming long entries.
3. Staleness scan: List memory files older than 30 days that have not been referenced. Recommend archival for files >90 days.
4. Frontmatter audit: Check that all memory files (except hot-cache.md) have name, description, and type in their frontmatter. Report any missing fields.

6. Save Results

Ask "Save these results for future sessions?" — if yes, write YYYY-MM-DD-<topic>.md to memory/. Add veto issues to memory/hot-cache.md only from auditor handoff or explicit user approval.

Examples, Advanced Features & Practical Limitations

Reference: See references/examples.md for three complete examples (hero keyword rankings, glossary lookup, e-commerce project init), advanced features (smart context loading, memory health check, bulk promotion/demotion, memory snapshot, cross-project memory, wiki lint), and practical limitations (concurrent access, cold storage retrieval, data freshness, wiki compilation).

GDPR / Privacy Compliance

memory/ may store third-party personal data — entity names, founder bios, LinkedIn profiles, author/journalist names surfaced by entity-optimizer or research skills. Under GDPR Art 4(1) (applies to processing of personal data of EU/EEA/UK residents regardless of where the controller is located), these qualify as "personal data". The user is the data controller. Non-EU users without EU/EEA/UK data subjects may still face analogous obligations under CCPA/CPRA (California), PIPEDA (Canada), LGPD (Brazil), or other national regimes. Not legal advice.

Retention policy

• WARM files: archive to memory/archive/ after 90 days unreferenced (default lifecycle)
• COLD archive: never auto-deleted, but eligible for Art 17 erasure requests
• All files: user MUST honor Art 17 requests from data subjects (individuals named in memory)

Deletion flow (Art 17 / CCPA §1798.105)

Invoke: memory-management purge <entity-name-or-slug>

This skill then:

1. Greps all files under memory/ (including memory/archive/) for the entity name, slug, or domain
2. Presents matches to user for confirmation
3. Deletes or anonymizes confirmed matches across canonical and derived surfaces:
   • Canonical: memory/hot-cache.md, WARM notes, COLD/archive files, memory/entities/<slug>.md, memory/entities/candidates.md, memory/geo-feedback/, audit aggregates, open loops
   • Wiki layer: compiled memory/wiki/ pages, memory/wiki/log.md, memory/wiki/log-archive/, memory/wiki/.unresolved.md, memory/wiki/.drift-log, memory/wiki/.retire-day-log
   • Phase 3 archive reverse-link scan (v9.9.9+): grep -l "originally_at:.*<entity>" memory/archive/*.md AND grep -l "retired_because_compiled:.*<entity>" memory/archive/*.md — purge entity name, originally_at path string, AND retired_because_compiled path string from any matched archive frontmatter
   • .unresolved.md value-field scrub (v9.9.9+): the .unresolved.md schema stores value_a: / value_b: as freeform values. If the entity name appears as a VALUE (e.g., value_a: "CEO is Jane Doe") rather than as the contradiction's entity: field, the standard entity-name grep catches it via line-match. Confirm by running grep -F "<entity-name>" memory/wiki/.unresolved.md AND scrubbing each match (replace value with redacted label, preserve the contradiction structure for audit integrity).
   • Malformed archive handling (v9.9.9+): some archives may lack originally_at: (legacy / manually-mv'd files) or contain non-printable bytes in the field. The reverse-link grep silently skips these, creating compliance ghosts. Mitigation: ALSO run grep -F "<entity-name>" memory/archive/*.md body-content grep — catches the entity name regardless of frontmatter integrity. If a malformed archive is touched by the purge, log it explicitly to memory/audits/gdpr-purges.md so a compliance audit can verify scope.
4. Removes or refreshes derived wiki indexes so purged names do not reappear from cached pages
5. Writes a tombstone to memory/privacy/tombstones.md with redacted label, salted non-reversible fingerprint, date, scope, and reingest_blocked: true; never store the raw subject
6. Logs the purge to memory/audits/gdpr-purges.md per the canonical schema in references/gdpr-purge-log-template.md (v9.9.9+) — required fields: purge_id, date, redacted_label, fingerprint, scope.{canonical,wiki,archive}, action, action_detail, legal_basis, proof.{grep_count_before,grep_count_after}, reingest_blocked: true, audit_signature. Auditor-verifiable structure: never raw subject; mechanical grep-count proof; cross-referenced to tombstone fingerprint.
7. For auditor archives, redact subject identifiers while preserving score/status/proof metadata required for audit integrity

Lawful basis reminder

Before writing a third-party person to memory/entities/, the user must have one lawful basis per GDPR Art 6 (where GDPR applies — see scope note above): consent, legitimate_interest, contract, or equivalent. Advisory — this skill does not enforce, and does not substitute for legal review.

Reference Materials

• references/examples.md — Worked examples, advanced features, practical limitations, and the auditor handoff archive block format & rules
• references/promotion-demotion-rules.md — Full promotion/demotion table and action procedures
• references/update-triggers-integration.md — Update procedures, archive routines, and cross-skill integration points
• CORE-EEAT Content Benchmark — Content quality scoring stored in memory
• CITE Domain Rating — Domain authority scoring stored in memory

Next Best Skill

Primary: keyword-research — seed or refresh campaign strategy with current demand signals.