agent-openai-memory
Add memory capabilities to your agent. Use when: (1) User asks about 'memory', 'state', 'remember', 'conversation history', (2) Want to persist conversations or user preferences, (3) Adding checkpointing or long-term storage.
62
73%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./agent-openai-advanced/.claude/skills/agent-memory/SKILL.mdQuality
Discovery
82%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a solid description with explicit trigger guidance and good keyword coverage. Its main weakness is that the 'what' portion is somewhat high-level ('add memory capabilities') without listing specific concrete actions like storing/retrieving memories, managing conversation buffers, or configuring persistence backends. The description also uses second person ('your agent') which is a minor style issue but not penalized as heavily as first person.
Suggestions
Add more specific concrete actions to the 'what' portion, e.g., 'Store and retrieve conversation history, manage user preference persistence, configure checkpointing for long-running agent sessions'
Narrow the trigger term 'state' to reduce conflict risk with unrelated state management skills, e.g., 'agent state' or 'conversation state'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain ('memory capabilities') and some actions like 'persist conversations', 'user preferences', 'checkpointing', 'long-term storage', but doesn't list multiple concrete specific actions (e.g., specific APIs, data structures, or operations). It's more about categories than concrete actions. | 2 / 3 |
Completeness | Clearly answers both 'what' (add memory capabilities to your agent) and 'when' with an explicit 'Use when:' clause listing three specific trigger scenarios. The when clause is well-structured with numbered conditions. | 3 / 3 |
Trigger Term Quality | Includes strong natural keywords users would say: 'memory', 'state', 'remember', 'conversation history', 'persist conversations', 'user preferences', 'checkpointing', 'long-term storage'. These cover a good range of terms a user might naturally use. | 3 / 3 |
Distinctiveness Conflict Risk | The terms 'state' and 'remember' are somewhat generic and could overlap with other skills (e.g., state management in UI frameworks, or general knowledge recall). However, the combination with 'memory', 'conversation history', and 'checkpointing' narrows the scope reasonably well. | 2 / 3 |
Total | 10 / 12 Passed |
Implementation
64%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill provides solid, actionable guidance with executable code examples and concrete configuration files for adding session-based memory to an OpenAI Agents SDK application on Databricks. Its main weaknesses are some unnecessary conceptual explanation that Claude doesn't need (the 'How Sessions Work' lifecycle description, the Key Concepts table) and a lack of explicit validation checkpoints in the setup workflow. The troubleshooting table is a nice addition but doesn't substitute for inline verification steps.
Suggestions
Remove or significantly condense the 'How Sessions Work' and 'Key Concepts' sections—Claude understands session patterns and doesn't need the conceptual overview or definition table.
Add explicit validation checkpoints in the setup workflow, e.g., 'Verify Lakebase instance is running: `databricks lakebase instances get <name>` before proceeding to configuration'.
Consider adding a brief verification step after configuration (e.g., 'Run `uv run start-app` and confirm no LAKEBASE_INSTANCE_NAME errors before testing multi-turn conversations').
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill includes some unnecessary explanatory content (e.g., 'How Sessions Work' section explaining the session lifecycle conceptually, the Key Concepts table restating obvious definitions). However, the code examples and configuration sections are reasonably tight. Could be tightened by removing the conceptual overview and table. | 2 / 3 |
Actionability | Provides fully executable code snippets for session creation, session ID extraction, configuration files (databricks.yml, .env), and curl commands for testing multi-turn conversations. All examples are copy-paste ready with clear placeholders. | 3 / 3 |
Workflow Clarity | The prerequisites and configuration steps are listed but lack explicit validation checkpoints. There's no 'verify your Lakebase instance is running before proceeding' step or validation between configuration and testing. The troubleshooting table partially compensates but feedback loops are missing from the setup workflow. | 2 / 3 |
Progressive Disclosure | References to other skills (lakebase-setup, run-locally, deploy) are present in 'Next Steps' but not clearly signaled throughout the document. The content is somewhat monolithic—the configuration files section and troubleshooting could potentially be separate references. However, for a skill of this size, the inline approach is borderline acceptable. | 2 / 3 |
Total | 9 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- databricks/app-templates
- Commit
1c88215
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.