LearnNewsExamplesServices
Frontmatter
id7334
titleDocument Human Verification of Agent Memory
stateClosed
labels
documentationenhancement
assigneestobiu
createdAtOct 3, 2025, 12:15 PM
updatedAtOct 3, 2025, 12:17 PM
githubUrlhttps://github.com/neomjs/neo/issues/7334
authortobiu
commentsCount0
parentIssue7316
subIssues[]
subIssuesCompleted0
subIssuesTotal0
blockedBy[]
blocking[]
closedAtOct 3, 2025, 12:17 PM

Document Human Verification of Agent Memory

Closed v11.0.0 documentationenhancement
tobiu
tobiu commented on Oct 3, 2025, 12:15 PM

To ensure the integrity of the agent's memory core, this ticket covers updating the .github/WORKING_WITH_AGENTS.md guide to instruct human developers on their role in the memory-saving process.

While the agent is mandated to follow a "save-then-respond" protocol, a sufficiently derailed agent might fail to do so. This update introduces the concept of a "human-in-the-loop" safeguard for the memory protocol itself.

Acceptance Criteria

  1. A new section, titled "The Memory Core: A Shared Responsibility", is added to .github/WORKING_WITH_AGENTS.md.
  2. This section explains the agent's "save-then-respond" duty.
  3. This section explicitly defines the human developer's duty to verify that the ai:add-memory tool call is successfully made after every agent turn when the memory core is active.
  4. The section provides a clear recovery prompt for the user to issue in case the agent fails to save its memory.
tobiu assigned to @tobiu on Oct 3, 2025, 12:15 PM
tobiu added the documentation label on Oct 3, 2025, 12:15 PM
tobiu added parent issue #7316 on Oct 3, 2025, 12:15 PM
tobiu added the enhancement label on Oct 3, 2025, 12:15 PM
tobiu referenced in commit 94a925d - "Document Human Verification of Agent Memory #7334" on Oct 3, 2025, 12:17 PM
tobiu closed this issue on Oct 3, 2025, 12:17 PM