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 BongSuCHOI/memex --skill remembering-conversationsgit clone --depth 1 https://github.com/BongSuCHOI/memexWrote 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/bongsuchoi/memex/remembering-conversations)<a href="https://agentmods.dev/skills/bongsuchoi/memex/remembering-conversations"><img src="https://agentmods.dev/badge/skills/bongsuchoi/memex/remembering-conversations/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/bongsuchoi/memex/remembering-conversations"><img src="https://agentmods.dev/badge/skills/bongsuchoi/memex/remembering-conversations.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 | $0.00048 | $0.00605 |
| Opus 5.5 | $0.00019 | $0.00242 |
| Sonnet 5.5 | $0.00010 | $0.00121 |
| Haiku 4.5 | $0.00005 | $0.00060 |
Grade A, and why
remembering-conversations 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 1mo 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 — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Remembering Conversations
Recover relevant evidence from Memex before repeating past work or inventing a
new rationale. These MCP calls are read-only. Their results are marked
memex_recall: they remain searchable context but are not eligible evidence for
new fact extraction merely because the agent repeated them.
Route the question
- Past implementation, discussion, or exact wording: call
search, thenreadonly the best 2–5 archive ranges. - Durable decision, preference, pattern, knowledge, or constraint: call
search_facts. - Why a fact exists: call
trace_fact; treat source exchanges as authoritative provenance and the separately labeled interpretive context as non-authoritative. - How decisions connect: call
explore_graph; usesearch_ontologyfor domain/category browsing. - How another project solved a similar problem: call
cross_project_insights. - A synthesis in the user's established style: call
ask_avatar, then retain its cited evidence and confidence boundary.
For project-sensitive tools, pass the current Codex thread's canonical absolute
cwd as project/current_project, or use an explicit global/all scope.
Never let the MCP server's installed-plugin cwd stand in for the user's project.
Memory tiers
A memory sits in one of three tiers: the branch tier (promotion_state
workstream/workspace), project-common (project-current/decision), or
global (scope_type: global). A project scope reads global plus
project-common plus the current branch tier, so a memory recorded on a different
branch will not appear until it is promoted. When a search comes back empty and
the user believes the memory exists, say that it may still be in another branch's
tier rather than asserting it was never recorded; trace_fact shows the tier and
any PROMOTED/DEMOTED event. Moving a memory between tiers is a user action —
memex facts promote|demote, or the promote/demote buttons in the Web UI memory
detail panel — never something to do on the user's behalf without being asked.
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.
- 1mo ago Changed · +13 lines 250864f85c8e
- 1mo ago Changed · +1 lines 9d2c0c6bab3b
- 1mo ago First seen · 40 lines · 48 tokens per session scan A 738539badb20
remembering-conversations is a skill published in the GitHub repository BongSuCHOI/memex (0 stars, last pushed 29d ago), licensed MIT. It adds 48 tokens to every session and 605 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
potpie-graph
Use when the task can read or write the project-memory graph through the potpie CLI: discover the contract with graph catalog, read named views with graph read, resolve entity identity with graph search-entities, create validated plans with graph propose, commit plans with graph commit --verify, inspect quality with…
potpie-repo-baseline
Use when establishing, refreshing, or deeply understanding a repository's baseline memory in Potpie: purpose, application type, features, services/modules, environments, deploy shape, dependencies, API contracts, datastores, integrations, ownership, and explicit preferences. The harness reads authored and…
potpie-debug-memory
Use while debugging or troubleshooting failures, flaky tests, incidents, production alerts, CI failures, local dev setup issues, repeated bugs, prior fixes, failed attempts, and verification history.
linglu-discipline
A set of 16 operating rules for an AI coding agent, including when to read or write long-term memory and how to report completed actions. It is designed to work with the Lingshu memory service through an MCP connection.
my-wiki
A skill for managing a local Markdown knowledge vault that stores captured webpages, PDFs, Office files, notes, images, folders, and ZIP files. It links preserved source material, called References, to durable topic pages, called Concepts.
hybrid-memory
Hybrid memory strategy combining OpenClaw's built-in QMD vector memory with Graphiti temporal knowledge graph. Use for all memory recall requests.