Skip to content
Agentmart
Submit a tool

memory-steward

Prepare, review, commit, verify, export, delete, or freeze durable agent memory under explicit consent and read-back control.

Skill by serhiipriadko2-sys

IndexedMIT
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/serhiipriadko2-sys-memory-steward

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Prepare, review, commit, verify, export, delete, or freeze durable agent memory under explicit consent and read-back control. Use when a user asks to remember, save, record, promote, forget, delete, export, or preserve context. Default to a non-persistent candidate. Bind consent to the exact candidate fingerprint, issuer, purpose, target, retention, and expiry; reject replay, secrets, sensitive personal data, direct Archive writes, and any success claim without connector read-back.

Use cases: data, security, agent-building

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/memory-steward

What's inside

Skill name
memory-steward
Description
Prepare, review, commit, verify, export, delete, or freeze durable agent memory under explicit consent and read-back control. Use when a user asks to remember, save, record, promote, forget, delete, export, or preserve context. Default to a non-persistent candidate. Bind consent to the exact candidate fingerprint, issuer, purpose, target, retention, and expiry; reject replay, secrets, sensitive personal data, direct Archive writes, and any success claim without connector read-back.
License
MIT
Instructions
13 lines in SKILL.md · view on GitHub

Files (4)

  • SKILL.md
  • openai.yaml · agents/
  • icon.svg · assets/
  • consent-contract.md · references/

SKILL.md (first lines)

# Memory Steward

1. Distinguish temporary conversation context from durable memory.
2. Produce a candidate containing exact content, purpose, sensitivity, retention, target, deletion path, and fingerprint.
3. Disclose the candidate before requesting consent.
4. Accept consent only from a trusted adapter and only when it matches the candidate fingerprint and remains unexpired.
5. Treat consent as one-time; reject replay or candidate substitution.
6. Write through the connector, read back the stored hash, then issue a receipt.
7. If any stage is unavailable, return the candidate and `memory write unavailable`; never invent persistence.

Apply [references/consent-contract.md](references/consent-contract.md).

From serhiipriadko2-sys/parallax, MIT.

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Jul 28, 2026 (3 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from serhiipriadko2-sys

Cite this page

memory-steward skill. Agent Mart. https://agentmart.dev/skills/serhiipriadko2-sys-memory-steward (last updated 2026-10-24)