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/byte5ai/palaia/palaia-memorynpx skills add byte5ai/palaia --skill palaia-memorygit clone --depth 1 https://github.com/byte5ai/palaiaWrote 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/byte5ai/palaia/palaia-memory)<a href="https://agentmods.dev/skills/byte5ai/palaia/palaia-memory"><img src="https://agentmods.dev/badge/skills/byte5ai/palaia/palaia-memory.svg" alt="Measured on agentmods" 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 | $0.00127 | $0.02576 |
| Opus 5 | $0.00063 | $0.01288 |
| Sonnet 5 | $0.00025 | $0.00515 |
| Haiku 4.5 | $0.00013 | $0.00258 |
Grade A, and why
palaia-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 6d 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 — 211 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Shared memory
This person keeps one memory that every assistant they use can read and write. It lives on their own machine as plain files, so anything you put in it is still there next week, in a different tool, in a different conversation.
Do this before you answer
Two checks, in this order, every turn — before you write your reply, not after:
- Did they just tell me something worth keeping? A decision, a reason, a
convention, a correction, a costly detail — even mentioned in passing on the
way to a different question. If yes,
captureit now. Answering first ends the turn and loses it. - Do I need to look something up? If the task touches their projects,
their conventions, or a name you cannot place,
recallbefore you decide.
Then answer, and mention in one short line what you saved.
Neither check is a favour to ask permission for. They are ordinary parts of doing the work, like reading a file before editing it. "Want me to save that?" is the wrong move: it spends the person's turn on something they already told you, and if they do not reply, it is gone.
The tools
The memory arrives as a family of tools whose names carry the memory they
belong to — work_memory_recall, personal_memory_capture, and so on. Use the
names your client actually lists; if two memories are mounted, the prefix is how
you tell them apart. Every tool description says what that memory is for.
The ones that matter here:
| Tool | Use it for |
|---|---|
recall |
What do we know about this topic? Ranked, most useful first. |
build_context |
Pick up where a previous session left off. |
search |
Find a specific note when you already know roughly what it is called. |
read |
Read one note in full. |
capture |
Save something you learned, and move on. |
write / edit |
Only when the person asked you to write or change a note. |
That last row is a hard line. Anything you learned goes in with capture —
never by writing a new note, and never by editing an existing one to bolt your
fact onto it. Those notes are the person's own, they may be read by ten other
sessions, and rewriting one to fit something you just heard is how a memory
quietly stops being trustworthy.
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.
- 6d ago First seen · 211 lines · 127 tokens per session scan A 96992b449a7f
palaia-memory is a skill published in the GitHub repository byte5ai/palaia (19 stars, last pushed today), licensed MIT. It adds 127 tokens to every session and 2,576 once invoked, about $0.0006 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 2026-08-30.
Other skills, from other repositories
setup
Set up or reconfigure session-recall for the user — install, pick the interaction language, choose a local or cloud embedder, optionally enable team sharing. Trigger when the user asks to "set up session-recall", "настрой session-recall", install it, change its embedder/language, or enable sharing with a colleague.…
memx
Use when an agent needs to recall, save, inspect, or audit memX memories through the bundled MCP tools.
session-recall
Search the unified local history of past Claude Code, Codex, and Cursor sessions at the START of a task when the user references a prior bug, feature, decision, file, or piece of work. Use before assuming fresh context for prompts such as "remember when…", "we worked on…", "the X bug", "back to…", "what did we decide…
mission-control
Interact with Mission Control — AI agent orchestration dashboard. Use when registering agents, managing tasks, syncing skills, or querying agent/task status via MC APIs.
memory-to-skill
Turn workflows from your MemSearch memory into reusable skills. Use when the user asks to make/create/extract/distill a skill from what they just did or from past work, review skill candidates, install a distilled skill, or 'turn this into a skill'. Manages MemSearch procedural-memory candidates under…
create-ex
你是一个帮助用户重建前任数字人格的助手。 你的目标是通过对话引导 + 微信聊天记录分析,生成一个能真实复现前任沟通风格和情感模式的 Persona Skill。.