CL

claude-memory-curator

Curate and clean up Claude Code instructions and memory files.

Install

mkdir -p .claude/skills/claude-memory-curator && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/17001" && unzip -o skill.zip -d .claude/skills/claude-memory-curator && rm skill.zip

Installs to .claude/skills/claude-memory-curator

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Audit, review, clean up, and prune Claude Code durable context. Use when the user asks about CLAUDE.md, CLAUDE.local.md, .claude/rules, user Claude rules, Claude Code auto memory, /memory, stale instructions, memory pollution, sensitive context, settings such as autoMemoryEnabled or claudeMdExcludes, or where a Claude instruction should live. Do not use for Codex memory, Cursor rules, Claude app memory, Anthropic API Memory Stores, or generic docs cleanup.
460 chars✓ has a “when” triggerlonger than Claude Code's old 250-char listing cap (fine on current versions)
Intermediate

Key capabilities

  • →Audit Claude Code durable context
  • →Review Claude Code instruction files
  • →Clean up stale or conflicting Claude instructions
  • →Propose better destinations for Claude context entries
  • →Scan for sensitive context in Claude memory

How it works

The skill inventories Claude context files, scans for risks like sensitive or stale entries, classifies atomic claims, and proposes actions based on predefined rules and precedence.

Inputs & outputs

You give it
User request to audit, review, clean up, or prune Claude Code memory or persistent instructions
You get back
A report of findings, proposed actions (e.g., KEEP, MOVE, DELETE), and a structured cleanup plan if requested

When to use claude-memory-curator

  • →Cleaning CLAUDE.md
  • →Pruning memory pollution
  • →Updating local instructions

About this skill

Claude Memory Curator

Goal

Audit Claude Code durable context as user-owned agent state: expose stale, unsafe, duplicated, over-broad, conflicting, misplaced, or unenforceable entries; propose better destinations; and route review, planning, persistence, and cleanup through one explicit contract.

Keep the subject scoped to Claude Code surfaces: CLAUDE.md, CLAUDE.local.md, conditionally loaded AGENTS.md, .claude/rules/, user-level Claude rules, settings, hooks, managed policy evidence, and auto memory files. Do not treat this as a Codex, Cursor, Claude app, or Anthropic API Memory Stores curator.

When to use

  • Use for review, placement, planning, or cleanup of the Claude Code durable surfaces named in the description.
  • Use when those surfaces are stale, conflicting, sensitive, over-broad, ignored, or causing Claude to reuse old guidance.

When not to use

  • Do not use for Codex, Cursor, Claude web/app, Anthropic API Memory Stores, generic prompt engineering, or repo-doc cleanup without Claude durable context.
  • Keep review requests read-only. Treat managed policy, including /etc/claude-code/CLAUDE.md and managed-settings.json, as read-only unless exact edit authority is explicit.

Workflow selection

Expose all eight workflows in this stable order, compactly when intent is clear. Recommend the route matching the requested outcome and delivery; for a bare invocation recommend review-chat while asking the user to choose:

WorkflowDelivery and result
plan-run-cleanup-fileOne record: review, approved plan, backup, cleanup, verification.
review-chatChat: read-only review and recommendations.
review-fileOne record: read-only review and recommendations.
cleanup-chatChat and backup: review, high-confidence atomic cleanup, verification.
cleanup-fileOne record and backup: review, high-confidence atomic cleanup, verification.
plan-cleanup-chatChat: review and approved plan; no cleanup.
plan-cleanup-fileOne record: review and approved plan; no cleanup.
plan-run-cleanup-chatChat and backup: review, approved plan, cleanup, verification.

Route from intent instead of adding an auto workflow:

  • Direct review defaults to review-chat; explicit persistence selects review-file.
  • Explicit cleanup without a delivery preference selects plan-run-cleanup-file.
  • A clear direct request may select another matching route when delivery and execution intent are explicit.
  • Agent-initiated activation may select only a relevant read-only route. Use review-file only when the existing task already requests persistence; never infer cleanup.
  • A bare invocation, conflicting cues, or ambiguity about review versus cleanup, chat versus file, execution, target paths, or mutation authority exposes the table and asks the user to choose.
  • A mutating route may be selected only when the user already requested cleanup of the identified Claude context scope.

Before substantive inspection, disclose the complete finite options once and state Selected, Reason, target paths, write scope, expected artifacts, protected state, Plan-mode capability, and unresolved approvals. Reuse facts and exact authority already established in the task; combine known fields into a short announcement. If selection is unambiguous, announce it and proceed. If it is ambiguous, stop before inventory and ask only the unresolved choice. Do not repeat the selector or ask for an unchanged selection again.

Workflow selection does not authorize whole-file deletion, destructive recovery, managed-policy edits, paid or external actions, deployment, publication, or scope expansion.

Inputs to inspect

  • When AGENTS.md is relevant, read references/context-surface-anatomy.md. Inventory presence separately from effective loading: confirm Claude version, provider/feature availability, settings, hook/plugin restrictions, and session loading evidence. Keep a working @AGENTS.md import for older or restricted hosts.
  • Resolve the repo, Claude home, configured auto-memory directory, and only the applicable project/user instruction, rule, setting, hook, managed-policy, and memory surfaces.
  • Inspect current repository evidence only as needed to verify a disputed claim.
  • Load the placement, classification, conflict, enforcement, and safe-editing references below only when their decision is active. Load the report or plan asset whenever producing that artifact.

Workflow

Every route applies the same review quality to the explicitly requested scope before planning or cleanup. Read the review procedure before substantive inspection. A single-file or single-claim request permits only targeted reads and the evidence needed to resolve an in-scope conflict; delivery never expands scope or reduces review quality.

Route execution

  • review-chat: return the review in chat and create no durable curation report.
  • review-file: persist the single curation record and make no context change.
  • cleanup-chat: derive only high-confidence atomic actions from the completed review, back up every exact file to be changed, apply them, re-read changed sections, and report verification in chat. Create no durable curation report.
  • cleanup-file: create the curation record before mutation; if persistence fails, stop. Then back up exact files, apply only high-confidence atomic actions, and complete the same record with execution and verification.
  • plan-cleanup-chat and plan-cleanup-file: enter the Plan lifecycle, resolve the cleanup plan with the user, and stop after approval without changing Claude context.
  • plan-run-cleanup-chat and plan-run-cleanup-file: enter the Plan lifecycle, resolve and approve the complete cleanup plan, recheck state, exit Plan mode, back up exact files, execute only the unchanged plan, and verify. Do not ask a generic second cleanup question after plan approval.

Direct cleanup (cleanup-chat or cleanup-file) is limited to high-confidence atomic edits, moves, or entry deletion in existing, editable, runtime-owned Claude context. Defer whole-file deletion, shared AGENTS.md, new context files, settings, hooks, managed policy, AGENTS.md, repository docs, skills, UI-only User/Team settings, uncertain schemas, medium/low-confidence changes, and any scope expansion. A plan-run route may execute broader curation changes only when the approved plan names each destination, write path, backup, rollback, and separate approval boundary.

Plan lifecycle

Respect active/requested native Plan mode and actual write permissions on every route: no report, backup, or cleanup write while Plan is active or permission is unknown. Planning may continue read-only without a manual mode switch. For plan-* routes or pending material questions, read references/plan-lifecycle.md. Reuse unchanged approval; mode exit alone is not approval, and unanswered questions grant no authority. Recheck state before writes and reconfirm only material changes. Plan-only routes never authorize cleanup.

Do not invoke claude-spec-interviewer inside this curation workflow. If findings require a broader durable rule, repository spec, or unresolved product decision, finish the selected curation route and offer the interviewer as a separate follow-up.

File delivery contract

File routes persist exactly one redacted curation record. Prefer an existing repository-native report location; otherwise use <repo>/.agent-reports/claude-memory-curation/<UTC timestamp>-<selection-id>.md. Create a new path without overwriting and keep all route output in that record.

The record contains Review, Plan, Execution Receipt, Deferred Work, Backup, and Verification. Use not applicable with a reason for phases the route does not perform. Create the record before mutation for cleanup-file and plan-run-cleanup-file; persistence failure blocks cleanup. Chat routes create no report file. Backup directories remain mandatory safety artifacts and do not count as curation reports.

Explicit --backup-root requires a stable non-sensitive --backup-root-alias; file routes persist the script-reported portable storage locator and <storage-locator>/backup-manifest.json. Report exact absolute backup and manifest paths only in non-persisted chat, never in repository artifacts.

Safety rules

  • Do not inspect when the route is unresolved, and do not mutate unless the selected route and user request authorize cleanup of the exact target scope.
  • Never silently delete, rewrite, truncate, or move Claude context files.
  • Back up every exact file before an approved edit and report the backup path.
  • Do not print full secrets, tokens, credentials, customer data, private identifiers, internal hostnames, or sensitive personal data.
  • If secret-like data is found, redact values in output, identify file and line when possible, recommend removal, and recommend rotation for real credentials.
  • If an auto-memory topic schema is unclear, do not edit it directly. Defer it in the current record or chat result.
  • Treat CLAUDE.md and auto memory as context, not enforcement. Recommend settings or hooks for deterministic restrictions.
  • Do not apply repo-specific assumptions globally. Prefer project CLAUDE.md, .claude/rules, AGENTS.md, or repo docs for repo rules.
  • Do not run broad destructive commands.

References

Read only when needed:

  • Placement and classification: references/context-surface-anatomy.md, references/classification-rubric.md.
  • Conflic

Content truncated.

When not to use it

  • →For Codex memory cleanup
  • →For Cursor rules or context cleanup
  • →For Claude web/app account memory or Anthropic API Memory Stores

Limitations

  • →Does not modify files when only review is requested
  • →Does not edit managed policy files by default
  • →Requires explicit approval for cleanup actions

How it compares

This skill specifically targets Claude Code's durable context, providing a structured approach to managing its unique instruction files and memory, unlike generic documentation or prompt engineering.

Compared to similar skills

claude-memory-curator side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
claude-memory-curator (this skill)04moReviewIntermediate
notion-knowledge-capture101yNo flagsIntermediate
feishu-doc148moNo flagsIntermediate
openspec-continue-change49moReviewIntermediate

Try saying

Example prompts that trigger this skill in your AI assistant.

Search skills

Search the agent skills registry