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 agentmods add skills/awrshift/claude-memory-kit/setupnpx skills add awrshift/claude-memory-kit --skill setupgit clone --depth 1 https://github.com/awrshift/claude-memory-kitWhat 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 | $0.00093 | $0.02642 |
| Opus 5 | $0.00046 | $0.01321 |
| Sonnet 5 | $0.00019 | $0.00528 |
| Haiku 4.5 | $0.00009 | $0.00264 |
Grade D, and why
setup scanned grade D with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
[ -f "$HOOK" ] || HOOK=$(find ~/.claude/plugins -path '*memory-kit*' -name session-start.py | head -1) Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
"Bash(rm -rf:*)" How it starts
The opening of the file, as written. The whole thing — 188 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/memory-kit:setup — adopt the kit in an existing repository
The plugin ships behaviour; the repository owns the state. This skill creates that state — always asking before the first write, and never touching files the repo already has.
Running without a user who can answer (a headless or -p run)? Then "offer" means skip and
report: take the documented default where one exists, install nothing that the skill marks as
opt-in, and finish with a list of what was deliberately not installed and what question is still
open. Never invent a preference on the user's behalf and never present a skipped step as done.
Step 0 — look before you write
Report in two lines what already exists: .claude/memory/MEMORY.md, context/handoffs/,
knowledge/, projects/, an existing documentation home (docs/, doc/, documentation/),
.claude/settings.json, .gitignore, and whether this repo already holds work of its own —
code, or documents other than a bare README (a real project) — or is effectively empty (a fresh
memory workspace). Being a git repo decides nothing; content does. Everything after this is a proposal the user
confirms once, not a sequence of prompts.
Step 1 — the shared memory layers
These four are shared across every project in the repo. Create only what is missing, copying
from ${CLAUDE_PLUGIN_ROOT}/templates/:
| Path | From | Purpose |
|---|---|---|
.claude/memory/MEMORY.md |
templates/MEMORY-TEMPLATE.md |
the hot cache the hook injects every session |
context/handoffs/HANDOFF-TEMPLATE.md |
templates/HANDOFF-TEMPLATE.md |
the per-session note format |
knowledge/index.md |
templates/workspace/knowledge-index.md |
the catalog of deep memory |
knowledge/concepts/.gitkeep |
— | where promoted patterns land |
.claude/state/.gitkeep |
— | hook bookkeeping (gitignored) |
Step 1b — the project layer (projects/<name>/)
The four layers above hold MEMORY. The work's own documents — tasks, specs, research, decisions,
QA — belong to a PROJECT, and the kit is multi-project by design: one folder per client or
product, so that a decision ledger, a findings registry and a QA protocol each count and describe
exactly ONE thing. This applies to a single-product code repository too — it simply has one
project folder, named after the product. (Earlier versions told you to skip projects/ in a code
repo. That left specs, backlogs and research with no home, and they scattered into the repo root.)
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.
- yesterday First seen · 188 lines · 93 tokens per session scan D e99c07778268
setup is a skill published in the GitHub repository awrshift/claude-memory-kit (31 stars, last pushed 5d ago), licensed MIT. It adds 93 tokens to every session and 2,642 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it D with 2 findings (reads agent configuration directories, recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
receipts
Show your personal model-quality trend card (failure rate, edit churn, tokens/task by model version).
provenance
Preview the provenance receipt (prompts, est. spend, tests run, agent-authored lines) for the current session ledger.
board
Show the wanted-poster bounty board: this repo's TODO/FIXME/HACK debt priced as aging XP bounties.
leaderboard
Show this repo's context-cost leaderboard: the files that pulled the most tokens into context.
dead-ends
List this repo's recorded dead ends: approaches already tried and abandoned, with reasons and cost.
scorecard
Show the CLAUDE.md compliance scorecard: which rules Claude actually follows vs ignores.