Skip to content

feat: include memory ID in store/update response text#146

Open
eisen0419 wants to merge 1 commit intoCortexReach:masterfrom
eisen0419:fix/issue-139-store-return-id
Open

feat: include memory ID in store/update response text#146
eisen0419 wants to merge 1 commit intoCortexReach:masterfrom
eisen0419:fix/issue-139-store-return-id

Conversation

@eisen0419
Copy link

Summary

  • Include the full memory ID in the response text of memory_store and memory_update tools
  • Prevents AI hallucination by providing verifiable confirmation of successful storage
  • Enables users and AI to reference specific memory entries by ID

Changes

  • memory_store: Added (ID: <uuid>) to response text
    • Before: Stored: "..." in scope 'agent:main'
    • After: Stored: "..." (ID: 8f82f232-...) in scope 'agent:main'
  • memory_update: Show full ID instead of truncated 8-char prefix
    • Before: Updated memory 8f82f232...: "..."
    • After: Updated memory (ID: 8f82f232-abc1-4def-...): "..."

Test plan

  • Verify memory_store response includes full UUID in text field
  • Verify memory_update response includes full UUID in text field
  • Confirm details.id field remains unchanged

Closes #139

Include the full memory ID in the response text of `memory_store` and
`memory_update` tools so the AI model can verify successful persistence
and reference specific entries.

- `memory_store`: add `(ID: <uuid>)` to response text
- `memory_update`: show full ID instead of truncated 8-char prefix

Closes CortexReach#139
@AliceLJY
Copy link
Collaborator

Codex static review summary:

Overall this looks low-risk and directionally correct. I did not find any security, resource, or logic bugs in the actual ID interpolation itself. memory_store and memory_update both now expose the full existing UUID, and details.id remains unchanged.

One compatibility note: this changes the user-visible content[].text contract for both tools. I did not find any in-repo parser depending on the old strings, but external agent prompts / scripts may still be pattern-matching on the previous text.

Suggestion before merge: add minimal regression coverage to lock the new success text for both memory_store and memory_update, and explicitly keep details.id unchanged.

My conclusion: approve with nits.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

After storing the new memory in the database, the returned message includes the memory storage ID

2 participants