remember

A session handoff tool that saves important project context to files and restores it in a later coding session.

In plain words
What is it for?
Saving session state at the end of work and reconstructing the current context when a new session begins.
Why use it?
New AI sessions start without memory, so important decisions and unfinished work can be lost. It checks the files on disk instead of trusting an earlier conversation.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/partikularwaters/banka/remember
Any agent
npx skills add partikularwaters/Banka --skill remember
Clone the repo
git clone --depth 1 https://github.com/partikularwaters/Banka

Made for: Claude Code, Codex.

Per session 54 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,454 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00054 $0.02454
Opus 5 $0.00027 $0.01227
Sonnet 5 $0.00011 $0.00491
Haiku 4.5 $0.00005 $0.00245

Measured 2d ago against content hash 4af19581e1fe, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

remember 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 2d 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.

skills-kit/remember/SKILL.md · 204 lines

How it starts

The opening of the file, as written. The whole thing — 204 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AI has no memory between sessions. Every new session starts blank. This skill fixes that — but the fix is the file on disk, never this conversation's own recollection of it, since another session may have touched the project since this conversation last looked.

Context Contract

Required: the resolved session-state file (tier-dependent — see "Resolve Banka state first" below) · current git log/status, checked before trusting this conversation · AGENTS.md/CLAUDE.md and the tier's required files, to resolve state.

Conditional: IDEA-SCOPE.md and the rest of the tier's Source-of-truth files, restore mode, when they exist · the resolved delegation queue, when present · a legacy CLAUDE.md chain, read-only, restore mode only.

Excluded by default: any file outside the tier's declared Source of truth · raw secret values, in any form, at any point.

Outputs: save mode — an updated session-state file, plus a one-line confirmation · restore mode — a conversational restore summary.

Write authority: save mode only — the resolved session-state file, by section, plus whichever file owns a globally-scoped fact a captured decision changes, plus the tier's overflow/ files and Overflow Index when the size thresholds above are crossed. Restore mode: none.

Resolve Banka state first

Before reading or writing project state, inspect AGENTS.md, the complete contents of CLAUDE.md, /core/, /context/, and the required tier files. Active schema 2 requires one complete Banka block in AGENTS.md containing these exact comments exactly once and in this order: <!-- BANKA:START -->, <!-- BANKA:STATE-SCHEMA: 2 -->, exactly one of <!-- BANKA:TIER: Minimal -->, <!-- BANKA:TIER: Core -->, or <!-- BANKA:TIER: Standard -->, then <!-- BANKA:END -->. The declared tier must match the filesystem shape and required files. CLAUDE.md must be exactly @AGENTS.md; if it is missing, schema 2 is still active for a runtime that discovers AGENTS.md directly, but report that Claude Code compatibility is unavailable.

Read the full file on GitHub · 204 lines

Changes

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.

  1. 2d ago First seen · 204 lines · 54 tokens per session scan A 4af19581e1fe

Subscribe to this mod's changes

remember is a skill published in the GitHub repository partikularwaters/Banka (3 stars, last pushed 3d ago), licensed MIT. It adds 54 tokens to every session and 2,454 once invoked, about $0.0003 per session on Opus 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.