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 kochetkov-ma/claude-brewcode --skill memory-syncgit clone --depth 1 https://github.com/kochetkov-ma/claude-brewcodeWrote 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/kochetkov-ma/claude-brewcode/memory-sync)<a href="https://agentmods.dev/skills/kochetkov-ma/claude-brewcode/memory-sync"><img src="https://agentmods.dev/badge/skills/kochetkov-ma/claude-brewcode/memory-sync/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/kochetkov-ma/claude-brewcode/memory-sync"><img src="https://agentmods.dev/badge/skills/kochetkov-ma/claude-brewcode/memory-sync.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 | $0.00037 | $0.00546 |
| Opus 5.5 | $0.00015 | $0.00218 |
| Sonnet 5.5 | $0.00007 | $0.00109 |
| Haiku 4.5 | $0.00004 | $0.00055 |
Grade A, and why
memory-sync 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 15d 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.
What it actually says
Memory sync
Sync project Codex instructions using references/sync-contract.md. Inventory concrete authorized paths before editing: applicable AGENTS.md, explicitly referenced .codex/rules/ or conventions, native agent TOML, and native skills/supporting instruction Markdown. Presence in a directory, git tracking, or being Markdown does not make a file authorized. User scope narrowing always wins, including "only rules" or an explicit file list.
Resolve the change-fact source from the request: session by default; explicit branch, commit/range, recent commits, or all facts. NORMAL verifies facts, deduplicates, and compacts; HARD additionally audits rule activation and removes generic guidance without losing project decisions. Depth never expands edit scope. Announce the source, depth, concrete inventory/exclusions, and verification in English.
Source code, distributed templates/product skill sources, public docs, config, reports, task/runtime state, caches, secrets, and personal/global instruction layers are evidence or excluded, never edit targets. Claude instructions and shared symlink targets are verify-only. Resolve real paths before editing; do not write through a symlink into another product or shared/global layer. Persistent Codex memory requires an explicit direct user request and the active memory policy; invoking this skill alone is insufficient authorization.
Snapshot the actual authorized files, then verify facts -> choose canonical homes/deduplicate -> compact with apply_patch. Preserve decisions, prohibitions, exceptions, ownership, provenance, numbered rule IDs, and supported native metadata. Re-audit included agents/skills for current roles, paths, tools, contracts, and references; parse TOML as TOML. Do not change operational behavior or permissions to make stale claims true.
Use bounded disjoint owners and an independent read-only post-sync check when authorized/available; otherwise disclose the verification limit. Verify every edit and deletion against the snapshot and live evidence, including the full text after large reductions. Keep the root rule index (Rule, Load when, Purpose) consistent with authorized rule edits: .codex/rules/ and paths: do not autoload. If the root index is outside the user's narrowed scope, report the required index correction rather than silently widening writes.
Report edited paths, verified facts, removed duplicates/stale claims, before/after size, uncertain facts, verify-only drift, checks, and deferred proposals. No automatic self-sync: this skill is editable only if explicitly included in the authorized inventory.
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.
- 15d ago First seen · 19 lines · 37 tokens per session scan A 68610a670e7c
memory-sync is a skill published in the GitHub repository kochetkov-ma/claude-brewcode (32 stars, last pushed today), licensed MIT. It adds 37 tokens to every session and 546 once invoked, about $0.0001 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 2026-10-01.
Other skills, from other repositories
brain-log
Show brain sync and evolution history.
Vizra ADK Memory System
Implement persistent memory, session context, and vector memory (RAG) for AI agents.
alive:session-history
Revive sessions (quick or heavy), browse, and search — 'what happened recently?', 'find the session where we discussed X', 'revive yesterday's session'. For single-session recall and multi-session browsing. If the human needs to merge multiple sessions into one working context or detect conflicts between parallel…
alive:mine-for-context
Deep context extraction from source material. Creates reference bundles, builds extraction plans, tracks what's been extracted, and discovers new targets — people, subjects, patterns, connections. The archaeologist that turns raw sources into structured knowledge. Can be invoked by alive:session-history for targeted…
alive-people
Weekly -- cross-reference people mentions, nudge stale contacts.
alive-capture
Capture external content into a walnut bundle -- emails, transcripts, screenshots, documents, research.