Persistent memory

Persistent Memory

Capture, recall, consolidate, and forget memories scoped to a user-chosen topic (e.g. parenting, work, relationship-anna) as plain markdown files, so any future conversation can pick up where the last one left off.

This SKILL.md is a thin index. Operation pipelines, taxonomy, privacy rules, integration patterns, and scaling guidance live in rules/*.md and load on demand. Literal artefact templates live in templates/*.md. Worked examples and citations live in references/*.md. Read only what the current operation asks for.

---

Mode Detection

Parse $ARGUMENTS (first token) and detect the operation:

Operation	Default	Trigger phrases
write	yes	"remember", "save to memory", "add to memory", $0 == "write"
read		"recall", "load memory", "what do you remember about", $0 == "read"
consolidate		"consolidate memory", "compress memory", $0 == "consolidate"
forget		"forget that", "delete memory", "redact", $0 == "forget"
list		"list memory", "what scopes do I have", $0 == "list"

State the detected operation and resolved scope in one line before continuing. Example:

Operation: write
Scope: parenting
Storage tier: home (~/.agent-memory/parenting/)

If no scope is provided, ask once (single batched message) — never guess.

---

Required Reading by Operation

Load on demand — do not preload.

Operation	Files
write	rules/storage-layout.md, rules/write-pipeline.md, rules/memory-taxonomy.md, rules/privacy-and-consent.md
read	rules/storage-layout.md, rules/read-pipeline.md
consolidate	rules/consolidate-pipeline.md, rules/memory-taxonomy.md
forget	rules/forget-pipeline.md, rules/privacy-and-consent.md
list	rules/storage-layout.md
integration	rules/integration-with-skills.md
scaling	rules/scaling-tiers.md
pre-flight	rules/quality-checklist.md, rules/anti-patterns.md

---

Storage Layout (one-line summary; full rules in rules/storage-layout.md)

Three tiers; the user picks per invocation, or accepts the default.

Tier	Path	Committed?	Default for
home (default)	~/.agent-memory/<scope>/	No	Personal scopes (parenting, work)
project-local	<repo>/.agent/memory/<scope>/	No (gitignore)	Per-project private notes
project-shared	<repo>/memory/<scope>/	Yes	Team-shared project knowledge

Per-scope directory layout (identical across tiers):

<storage-root>/<scope>/
├── INDEX.md          # Curated, ≤ 200 lines; always loaded by `read`
├── entries/          # Individual memory entries; loaded on demand
│   └── <yyyy-mm-dd>-<slug>.md
├── archive/          # Consolidated / superseded entries (audit trail)
└── AUDIT.log         # Append-only ledger of write / consolidate / forget

---

Core Workflow

Every operation is gated. Do not proceed to the next phase until the prior phase's gate passes.

write (default)

Phase	Name	Rule	Gate
0	Resolve scope + tier	rules/storage-layout.md	Scope name + storage tier confirmed; directory created
1	Privacy pre-flight	rules/privacy-and-consent.md	No secrets / PII on the never-store list slip through
2	Extract candidates	rules/write-pipeline.md	Candidate list produced with type, confidence, source per item
3	Compare to existing	rules/write-pipeline.md	Each candidate tagged ADD / UPDATE / DELETE / NOOP
4	Consent preview	rules/privacy-and-consent.md	User saw the diff and approved (unless --auto flag)
5	Write + audit	rules/write-pipeline.md	INDEX updated, entry files written, AUDIT.log line appended

read

Phase	Name	Rule	Gate
0	Resolve scope	rules/storage-layout.md	Scope directory exists; INDEX.md present (or report empty)
1	Load INDEX	rules/read-pipeline.md	INDEX content surfaced to current conversation
2	On-demand fetch	rules/read-pipeline.md	Detail entries fetched only when INDEX points to them

consolidate

Phase	Name	Rule	Gate
0	Snapshot	rules/consolidate-pipeline.md	Pre-consolidation state captured (path + file count)
1	Group + merge	rules/consolidate-pipeline.md	Semantically similar entries grouped; merge plan drafted
2	Prune stale	rules/consolidate-pipeline.md	Entries past staleness cutoff flagged for archive
3	Preview + apply	rules/consolidate-pipeline.md	User saw before / after summary and approved
4	Rewrite INDEX	rules/consolidate-pipeline.md	INDEX reflects new state; AUDIT.log appended

forget

Phase	Name	Rule	Gate
0	Resolve target	rules/forget-pipeline.md	Memory id, slug, or query resolves to exactly one entry set
1	Show + confirm	rules/forget-pipeline.md	User saw the entries and explicitly confirmed
2	Delete or redact	rules/forget-pipeline.md	Entries removed (or redacted); INDEX + AUDIT.log updated

list

Walk every storage tier the user has enabled, print every scope with entry counts and last-updated timestamps. No writes.

---

Integration With Other Skills

This skill is model-invocable (disable-model-invocation: false) so host workflows can call it programmatically. Two ways to invoke it:

1. Explicit — the user types /persistent-memory write parenting or /persistent-memory read parenting.
2. Runtime, from a host skill — the host skill's SKILL.md contains a one-line pointer block that calls Skill("persistent-memory", "read <scope>") when the host runs.

The second form is the canonical integration. Runtime Skill() calls require disable-model-invocation: false — without it the Skill tool refuses the call at the harness layer (you'd see Skill X cannot be used with Skill tool due to disable-model-invocation). See rules/integration-with-skills.md for the full contract and the literal snippet at templates/pointer-snippet.md.

Pattern	Token cost	Magic	Best for
Pointer	INDEX only, on skill load	None	The default. Explicit, debuggable, no hook.
Hook	INDEX every session	High	Always-on scopes (e.g. a personal assistant).

For the parenting example: add one block to parenting/SKILL.md:

> **Persistent memory:** Before responding, run
> `Skill("persistent-memory", "read parenting")` to load accumulated
> context for this scope.

---

Scaling Tiers (full guidance in rules/scaling-tiers.md)

Tier	Backend	Use when
1	Plain markdown (this skill, default)	≤ ~500 entries per scope, single user, no semantic search needed
2	Markdown + SQLite FTS index (this skill, opt-in)	Up to ~5k entries per scope, keyword search beats full-INDEX scan
3	Markdown blobs + local vector DB (Chroma, Qdrant)	Semantic recall ("what did we discuss about X") matters
4	Managed memory layer (Mem0, Letta, Zep)	Multi-user, multi-tenant, > 10k entries, graph relationships, hosted SLA

Graduate one tier at a time. The skill ships a migration recipe in rules/scaling-tiers.md for moving from markdown to SQLite, and from SQLite to a vector DB, without losing entries.

---

Core Principles

1. Plain text, local-first. Memory is markdown the user can read, edit, grep, and delete with standard tools. No proprietary format.
2. Progressive disclosure. INDEX is small and always loaded; detail entries load on demand. Modeled on Claude Code's MEMORY.md.
3. Two-phase write. Extract candidates first, then resolve each against existing entries with ADD / UPDATE / DELETE / NOOP. Modeled on Mem0's extraction + update pipeline.
4. Consent before persistence. Every write shows the user a diff preview unless --auto is passed; secrets and PII on the never-store list are refused outright.
5. Forgetting is a feature. A clear forget operation is part of the surface, not an afterthought. Required for privacy and for pruning entrenched mistakes (see Reflexion entrenchment warning).
6. One scope, one purpose. Resist mega-scopes ("life"). Split into parenting, health, work etc. so the INDEX stays under 200 lines.
7. Markdown until it hurts. Stay on Tier 1 until a concrete signal (search latency, INDEX bloat, multi-user) forces a graduation.

---

Anti-patterns (one-liner — full list in rules/anti-patterns.md)

• Writing to memory without showing the user the diff first.
• Storing secrets, credentials, or government IDs (refuse outright).
• Mega-scopes that swell the INDEX past 200 lines.
• Letting the model auto-write without scope confirmation.
• Re-recording the same fact instead of UPDATE-ing the existing entry.
• Deleting an entry without an AUDIT.log line.
• Skipping consolidation forever — the INDEX rots and recall degrades.
• Committing ~/.agent-memory/ to a public repo.

---

Definition of Done

A write run is done when:

• Scope and storage tier explicitly resolved (no defaults assumed silently).
• Privacy pre-flight passed (no never-store items in candidates).
• Candidate list shown to the user with ADD / UPDATE / DELETE / NOOP tags.
• User approved (or --auto flag was explicit).
• Entry files written; INDEX updated; AUDIT.log line appended.
• One-line summary delivered: "Saved N memories to <scope> (<tier>)."

A read run is done when:

• Scope resolved; if no INDEX exists, user is told the scope is empty.
• INDEX content is now in the conversation context.
• Detail entries are fetched only when the INDEX references them.

A consolidate run is done when:

• Before / after summary shown (entry count delta, INDEX line delta).
• User explicitly approved.
• AUDIT.log line appended with operation, timestamp, and counts.

A forget run is done when:

• Target entries shown verbatim before deletion.
• User confirmed (typed "yes" or --confirm flag).
• Entries removed (or redacted); INDEX updated; AUDIT.log line appended.