New: 97,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Memory Clean

by hashgraph-online in hashgraph-online/awesome-codex-plugins

A skill your agent uses when the user explicitly asks to clean, compact, prune, trim, deduplicate, optimize, or reduce Wingman memory, or asks to resolve stale or conflicting memory rules.

Apache-2.0Auto-check passed

Install Memory Clean

skills CLI
$ npx skills add hashgraph-online/awesome-codex-plugins --skill memory-clean -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install hashgraph-online/awesome-codex-plugins memory-clean --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/hashgraph-online/awesome-codex-plugins.git skills-src && mkdir -p .claude/skills && cp -r skills-src/plugins/lsshym/wingman.ai/skills/memory-clean .claude/skills/memory-clean && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
memory-clean
GitHub stars
1.2k
Token cost
~1.6k tokens
SKILL.md length
774 words
Files
4 (incl. scripts)
Skills in repo
736
Repo updated
First seen
Licence
Apache-2.0

At a glance

A skill your agent uses when the user explicitly asks to clean, compact, prune, trim, deduplicate, optimize, or reduce Wingman memory, or asks to resolve stale or conflicting memory rules.

  • Works in 5 steps: Check .wingman/memory/. → Check .wingman/memory/brief.md and… → If core files are missing, stop and… → …
  • The user explicitly asks to clean
  • SKILL.md covers Ownership Model, Repository Gate, Scope Selection and Candidate Types, plus 4 more sections
  • Runs Shell scripts from its folder

What it does

Memory Clean is an agent skill from hashgraph-online/awesome-codex-plugins. Use when the user explicitly asks to clean, compact, prune, trim, deduplicate, optimize, or reduce Wingman memory, or asks to resolve stale or conflicting memory rules.

Its SKILL.md is about 1.6k tokens, which your agent loads only when the skill is triggered. The skill folder holds 5 other files, including scripts (for example `examples/deletion-proposals.md`, `examples/lossless-compaction.md` and `scripts/memory-stats.sh`).

The repository describes itself as: A curated list of awesome OpenAI Codex / ChatGPT plugins, skills, and resources. The 1 Codex Marketplace. See live plugins at: https://hol.org/plugins/best-codex-plugins. The licence is Apache-2.0.

When your agent uses it

  • The user explicitly asks to clean
  • Reduce Wingman memory
  • Asks to resolve stale
  • Conflicting memory rules

Example prompts

  • “/memory-clean”

Requirements

  • A Bash shell

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

  1. Check .wingman/memory/.
  2. Check .wingman/memory/brief.md and .wingman/memory/context.md.
  3. If core files are missing, stop and report repository memory state.
  4. Read brief.md and context.md first.
  5. Read domain or history files only when the requested cleanup scope points to them.

What it can do on your machine

Read from SKILL.md and the folder at commit 16b4156. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 1 file in scripts/ (Shell), which the agent can run.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Memory Clean loads about 1.6k tokens when it runs. Until then it costs about 45 tokens; SKILL.md has 774 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~45
When it runs · the whole SKILL.md, loaded when a task matches
~1.6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from hashgraph-online/awesome-codex-plugins at commit 16b4156, republished under its Apache-2.0 licence (© hashgraph-online). 774 words, ~1,632 tokens.

Download SKILL.mdSave it as .claude/skills/memory-clean/SKILL.md (or your agent's skills folder). This skill also uses 3 other files; get the full folder from GitHub.
name
memory-clean
description
Use when the user explicitly asks to clean, compact, prune, trim, deduplicate, optimize, or reduce Wingman memory, or asks to resolve stale or conflicting memory rules.

Wingman Memory Clean

Clean Wingman memory only when the user explicitly asks. Preserve the ownership model while reducing default-read noise and conflicts.

Ownership Model

text
brief.md / domains/ = current projection, current binding truth bodies
history/events/     = event log, historical event bodies
history indexes     = projection indexes for historical lookup
context.md          = hot cache, active work state and short pointers

Cleanups should move memory toward one owning body per durable item:

  • current rule body lives in brief.md or domains/;
  • historical event body lives in history/events/;
  • historical lookup entry lives in projection indexes;
  • hot state or short pointer lives in context.md.

Repository Gate

  1. Check .wingman/memory/.
  2. Check .wingman/memory/brief.md and .wingman/memory/context.md.
  3. If core files are missing, stop and report repository memory state.
  4. Read brief.md and context.md first.
  5. Read domain or history files only when the requested cleanup scope points to them.

Never scan all memory files by default.

Scope Selection

Use the smallest scope that matches the request:

ScopeReadUse When
contextcontext.md, relevant current truth/history if linkedContext contains stale hot state, verbose old logs, or duplicate bodies.
current-truthbrief.md, relevant domainsCurrent rules are duplicated, stale, conflicting, missing identity, or missing relations.
history-indexrelevant projectionsHistory projections are bloated, missing useful routing, or copying event bodies.
history-eventnamed event bodies onlyUser explicitly asks to inspect or edit specific history events.
migrationtarget old context logs plus needed destinationsOld context logs need conversion into current truth, history, or pointers.
delete-proposaltarget files onlyUser asks to delete logs or remove noise.

If unclear, choose context unless the user mentions current rules, domains, Project Decisions, or history.

Candidate Types

Clean only when at least one concrete candidate exists:

  • COMPACT_TO_POINTER: context has details already owned by current truth or history.
  • PROMOTE_CURRENT_TRUTH: context contains a durable rule future agents must obey.
  • PROMOTE_HISTORY: context contains a durable change narrative or reasoning event.
  • PROMOTE_BOTH: context contains both current rule and historical explanation.
  • SUPERSEDE: an old current rule was replaced and should become superseded or deprecated.
  • REPAIR_RELATION: same-subject entries need updates, extends, or conflict resolution.
  • REPAIR_INDEX: history projection should be link-only or needs a domain/topic route.
  • DELETE_CANDIDATE: duplicate, obsolete, sensitive, or safely represented elsewhere.
  • NO_ACTION: cleanup cost or ambiguity exceeds benefit.

File size and line count are diagnostics only. They are not cleanup triggers by themselves.

Retention Review

Before changing a candidate, decide what would be lost:

  • Is it current truth, hot state, or history?
  • Is it still valid, replaced, or obsolete?
  • Is it the only evidence explaining a current rule?
  • Is a future task likely to need the rule, reason, failure mode, or pointer?
  • Can the same meaning be recovered from a current truth ID or history event link?

If uncertain, prefer NO_ACTION or a deletion proposal over irreversible cleanup.

Valid Actions

Compact Context

Replace verbose context with a short pointer only after the durable meaning is preserved elsewhere.

Valid pointer shape:

md
- <one-line state>; current truth: `<id/path or None>`; event: `<path or None>`; next: <immediate action/blocker or None>.

Preserve pending tasks, active blockers, and unresolved current work.

Show full SKILL.md (318 more words)Show less
Promote Current Truth

Write durable rules to brief.md or relevant domains/ when future agents must obey them. Include stable ID, Subject, Status, Rule, Applies When, Evidence, Confidence, Relation, Since, and History.

Promote History

Write historical events under history/events/YYYY/MM/ when the old context contains durable rule evolution, old-to-new meaning, important correction, migration, incident, regression fix, or user-requested historical memory. Update only needed projections.

Repair Current Rules

For A -> B requirement changes, make B the only current rule for overlapping scope, mark A as superseded or deprecated, and link with updates <old-id>.

If both rules remain valid, link with extends. If they conflict and evidence does not settle the winner, stop and ask.

Repair History Projections

Projection indexes should contain short summaries and links only. Remove copied event bodies from indexes only when the event body remains linked.

Delete

Never delete without explicit user confirmation after showing a proposal with exact IDs, paths, reasons, preserved locations, and risks.

Do not delete:

  • pending tasks;
  • current work;
  • unresolved bugs;
  • current Project Decisions;
  • current domain truths;
  • user-protected notes;
  • the only evidence explaining a current rule;
  • history event bodies, unless the user explicitly asked to delete those specific records.

Optional Resources

  • Run scripts/memory-stats.sh when heading layout, file size, or candidate discovery is unclear. The script is read-only.
  • Read examples/lossless-compaction.md when compaction safety is unclear.
  • Read examples/deletion-proposals.md when preparing a deletion proposal.

Do not read examples or run scripts by default.

Workflow

  1. Apply Repository Gate.
  2. Select the smallest cleanup scope.
  3. Identify concrete candidates.
  4. Run Retention Review.
  5. Classify candidates.
  6. Apply safe COMPACT_TO_POINTER, PROMOTE_CURRENT_TRUTH, PROMOTE_HISTORY, PROMOTE_BOTH, SUPERSEDE, REPAIR_RELATION, and REPAIR_INDEX changes.
  7. For DELETE_CANDIDATE, present a proposal and wait for explicit confirmation.
  8. Report files read, files changed, compactions, promotions, superseded rules, repaired indexes, and proposed or confirmed deletions.

Do not say memory was cleaned if no file changed. If only a deletion proposal was produced, say cleanup is pending confirmation.

© hashgraph-online, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 3 other files (scripts) in plugins/lsshym/wingman.ai/skills/memory-clean of hashgraph-online/awesome-codex-plugins.

  • SKILL.md
  • examples/deletion-proposals.md
  • examples/lossless-compaction.md
  • scripts/memory-stats.sh

Open the folder on GitHubat commit 16b4156

Compare with similar skills

Memory Clean next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Memory Clean compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Memory Clean this skillhashgraph-online/awesome-codex-plugins1.2k—~1.6kAutomated safety check: PassApache-2.0
Strategic Compactaffaan-m/ECC274k1 repos~2.1kAutomated safety check: PassMIT
No Explicit Anythedaviddias/Front-End-Checklist74k—~565Automated safety check: PassMIT
Android Clean Architectureaffaan-m/ECC274k4 repos~2.2kAutomated safety check: PassMIT
Compactcatlog22/Claude-Code-Workflow2.1k—~3.2kAutomated safety check: PassMIT
Strategic Compactaffaan-m/ECC274k2 repos~434Automated safety check: PassMIT

Similar skills

  • Strategic Compact

    affaan-m/ECC

    Suggests manual context compaction at logical intervals to preserve context through task phases rather than arbitrary auto-compaction.

    274k GitHub starsUsed in 1 repo~2.1k tokens
    Agent WorkflowsAuto-check passed

  • No Explicit Any

    thedaviddias/Front-End-Checklist

    A skill your agent uses when reviewing TypeScript files for type safety regressions, during code review of functions that handle external data, or when the codebase has ESLint warnings for…

    74k GitHub stars~565 tokensUpdated yesterday
    DevelopmentAuto-check passed

  • Android Clean Architecture

    affaan-m/ECC

    Applies Clean Architecture to Android and Kotlin Multiplatform projects: module layout, dependency rules, UseCases, Repositories and data layer patterns.

    274k GitHub starsUsed in 4 repos~2.2k tokens
    DevelopmentAuto-check passed

  • Compact

    catlog22/Claude-Code-Workflow

    Compact current session memory into structured text for session recovery.

    2.1k GitHub stars~3.2k tokensUpdated 3 mo ago
    Agent WorkflowsAuto-check passed

  • Strategic Compact

    affaan-m/ECC

    任意の自動コンパクションではなく、タスクフェーズを通じてコンテキストを保持するための論理的な間隔での手動コンパクションを提案します。

    274k GitHub starsUsed in 2 repos~434 tokens
    Auto-check passed

  • Strategic Compact

    affaan-m/ECC

    임의의 자동 컴팩션 대신 논리적 간격에서 수동 컨텍스트 압축을 제안하여 작업 단계를 통해 컨텍스트를 보존합니다.

    274k GitHub starsUsed in 2 repos~724 tokens
    Auto-check passed

More from hashgraph-online/awesome-codex-plugins

All 736 skills in this repo

  • Anime Reaction Gif

    hashgraph-online/awesome-codex-plugins

    Create original anime-style reaction stickers as looping GIFs and MP4 previews, using generated character pose sheets and timed key poses.

    1.2k GitHub stars~922 tokensUpdated yesterday
    Auto-check passed

  • Calibredb

    hashgraph-online/awesome-codex-plugins

    Manage and query Calibre libraries with the calibredb CLI (local paths or Calibre Content server URLs).

    1.2k GitHub stars~1k tokensUpdated yesterday
    Auto-check passed

  • Rust API Test Harness

    hashgraph-online/awesome-codex-plugins

    A skill your agent uses when adding, changing, testing, or debugging Rust HTTP APIs and services, especially when Codex needs black-box integration tests, random-port app startup, real database test…

    1.2k GitHub stars~1.7k tokensUpdated yesterday
    Auto-check passed

  • Art

    hashgraph-online/awesome-codex-plugins

    Make a studio's game look like something at build time — a cover from a real frame of the game (free), painted covers, backdrops, textures and character plates from image models through the…

    1.2k GitHub stars~2.4k tokensUpdated yesterday
    Auto-check passed

  • Calle

    hashgraph-online/awesome-codex-plugins

    Use CALL-E from Codex through the calle CLI. An agent skill from hashgraph-online/awesome-codex-plugins.

    1.2k GitHub stars~2.9k tokensUpdated yesterday
    Auto-check passed

  • Game Balance Economy

    hashgraph-online/awesome-codex-plugins

    Balance game difficulty, resources, rewards, probability, progression, economies, and dominant strategies.

    1.2k GitHub stars~618 tokensUpdated yesterday
    Auto-check passed

Questions about Memory Clean

What does Memory Clean do?

A skill your agent uses when the user explicitly asks to clean, compact, prune, trim, deduplicate, optimize, or reduce Wingman memory, or asks to resolve stale or conflicting memory rules. Memory Clean is an agent skill from hashgraph-online/awesome-codex-plugins. Use when the user explicitly asks to clean, compact, prune, trim, deduplicate, optimize, or reduce Wingman memory, or asks to resolve stale or conflicting memory rules.

When should I use Memory Clean?

Memory Clean fits situations like: the user explicitly asks to clean; reduce Wingman memory; asks to resolve stale; conflicting memory rules.

How do I install Memory Clean in Claude Code?

Run `npx skills add hashgraph-online/awesome-codex-plugins --skill memory-clean -a claude-code`. Or copy the skill folder (plugins/lsshym/wingman.ai/skills/memory-clean in hashgraph-online/awesome-codex-plugins) into .claude/skills/memory-clean in your project. Claude Code loads it when a task matches its description.

How do I install Memory Clean in Codex?

Run `npx skills add hashgraph-online/awesome-codex-plugins --skill memory-clean -a codex`. Or copy the skill folder (plugins/lsshym/wingman.ai/skills/memory-clean in hashgraph-online/awesome-codex-plugins) into .agents/skills/memory-clean in your project. Codex loads it when a task matches its description.

Can I use Memory Clean in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add hashgraph-online/awesome-codex-plugins --skill memory-clean -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/memory-clean, .gemini/skills/memory-clean, .github/skills/memory-clean and .opencode/skills/memory-clean in your project.

What does Memory Clean need to run?

Going by SKILL.md and its folder, Memory Clean needs a shell for the scripts in its folder. Our summary lists: A Bash shell.

Does Memory Clean access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Memory Clean safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Memory Clean use?

Memory Clean is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Memory Clean use?

About 1.6k tokens (SKILL.md is roughly 6.5k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Memory Clean?

Skills that share tags, products or a category with Memory Clean: Strategic Compact (affaan-m/ECC, 274k stars), No Explicit Any (thedaviddias/Front-End-Checklist, 74k stars), Android Clean Architecture (affaan-m/ECC, 274k stars) and Compact (catlog22/Claude-Code-Workflow, 2.1k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Memory Clean?

hashgraph-online (a GitHub organization) maintains it in hashgraph-online/awesome-codex-plugins, which has 1,232 GitHub stars. The repository holds 736 skills in this directory. The repository was last updated on October 6, 2026.

Source: hashgraph-online/awesome-codex-plugins on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.