Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Heapy/kortex --skill clean-claude-memorygit clone --depth 1 https://github.com/Heapy/kortexWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/heapy/kortex/clean-claude-memory)<a href="https://agentmods.dev/skills/heapy/kortex/clean-claude-memory"><img src="https://agentmods.dev/badge/skills/heapy/kortex/clean-claude-memory/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/heapy/kortex/clean-claude-memory"><img src="https://agentmods.dev/badge/skills/heapy/kortex/clean-claude-memory.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00047 | $0.00718 |
| Opus 5.5 | $0.00019 | $0.00287 |
| Sonnet 5.5 | $0.00009 | $0.00144 |
| Haiku 4.5 | $0.00005 | $0.00072 |
Grade A, and why
clean-claude-memory scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 27d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Clean Claude Memory
Claude Code stores project-scoped durable notes outside the repository:
~/.claude/projects/<encoded-absolute-project-path>/memory/
MEMORY.md is the index. Other Markdown files are topic memories. Memory is a hint for future sessions, not an authoritative source; verify every claim against current code, docs, configuration, tools, and project state.
Treat recorded plans as historical artifacts only, not as documentation or a source of truth. Never use them to implement other features or as authoritative evidence about current behavior; verify relevant claims against current code, tests, configuration, requirements, and project documentation.
Do not include these unless the user explicitly expands the scope:
- repository
CLAUDE.mdorAGENTS.md; - global
~/.claude/CLAUDE.md; - memories belonging to another project.
Workflow
- Resolve the current project's memory directory.
- Read
MEMORY.mdand inventory every sibling memory file. - Review topic files one at a time. For each file:
- read it completely;
- explain what it preserves;
- verify its claims against current authoritative sources;
- identify stale facts, duplication, historical narration, and transient paths, versions, plans, or commands;
- recommend one action;
- ask the user one direct question and wait before reviewing the next file.
Use these actions:
| Action | Use when |
|---|---|
| Keep | Unique, durable intent or preference remains accurate and useful. |
| Rewrite | A valuable core remains, but the file contains stale or excessive detail. |
| Delete | The content is obvious, obsolete, historical, or already authoritative elsewhere. |
| Move | The information belongs in code, tests, docs, a backlog task, or a reusable skill. |
For Move, create and verify the new home before deleting the memory:
- code invariant or non-obvious trade-off → nearby comment or test;
- actionable future work → project task or issue;
- durable project contract → project documentation;
- reusable agent workflow → skill.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 27d ago Changed · +1 lines 70bd696c8ff1
- 1mo ago First seen · 79 lines · 47 tokens per session scan A ed072b714edf
clean-claude-memory is a skill published in the GitHub repository Heapy/kortex (7 stars, last pushed yesterday), licensed Apache-2.0. It adds 47 tokens to every session and 718 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
codex-history-recovery
A recovery tool for Codex Desktop conversation history when chats become hidden after changing accounts, API keys, providers, or models.
learning-memory
Use when capturing or restoring a learner's persistent profile to personalize teaching across sessions.
codex-session-history
List and inspect local Codex session history by reading /.codex/sessionindex.jsonl, /.codex/sessions/, and /.codex/archivedsessions/. Default to unarchived sessions in /.codex/sessions/. Use when users ask what Codex sessions exist, want each session's id or title, need to know which project or workspace a session…
principle-guard-the-context-window
Apply when context is filling up: large outputs, long files, repeated reads, fan-out planning. Route bulk to subagents; keep summaries in the main thread, not raw payloads.
promote-memory
Review candidate learnings in Claude Code's native auto memory (/.claude/projects/ /memory/, machine-local) and run them through a five-critic council in parallel: generality, staleness, redundancy, evidence, format. Majority vote (3+ of 5) promotes the entry to MEMORY.md. Use when user says "promote memory", "review…
compress-session
Distill the current conversation into a structured note (decisions made, open questions, file pointers with line numbers, next 1–3 actions) and save to qualityreports/sessionlogs/ before auto-compression. Differs from /checkpoint (explicit stop-point snapshot) and from auto-compaction (which truncates rather than…