Name: modify-agent
Rating: 68 (1 reviews)
Author: databricks

modify-agent

Modify agent code, add tools, or change configuration. Use when: (1) User says 'modify agent', 'add tool', 'change model', or 'edit agent.py', (2) Adding MCP servers to agent, (3) Changing agent instructions, (4) Understanding SDK patterns.

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

The canonical home for this skill is modify-agent in databricks/app-templates

Quality

Content

71%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a dense, actionable reference with strong code examples and useful error-avoidance guidance for customizing agent instructions. Its main weakness is progressive disclosure: with no bundle files present, substantial API-reference content is inlined into SKILL.md instead of being moved into a separate reference file and summarized here.

Suggestions

Move the per-class SDK API reference (ChatDatabricks, DatabricksEmbeddings, DatabricksVectorSearch, and the MCP server variants) into a bundled reference file (e.g. references/sdk-api.md) and keep SKILL.md as a concise overview with a one-line pointer to it.

Add an inline post-edit verification step (e.g. 'after changes, run the agent locally and confirm it streams a response') rather than deferring all testing to the run-locally skill.

Trim or relocate the time-sensitive model-name list so specific endpoint names don't risk staleness inside the main skill body.

Dimension	Reasoning	Score
Conciseness	The body is code-forward and largely efficient, avoiding explanations of concepts Claude already knows (it does not define MCP, LangGraph, or MLflow); minor trim opportunities remain, such as the time-sensitive model-name list and the verbose Genie-space ID extraction example.	4 / 5
Actionability	Provides mostly executable, copy-paste-ready code across ChatDatabricks, embeddings, vector search, five MCP server variants, and create_agent usage, with minor gaps such as process_agent_astream_events shown only via call site rather than full signature.	4 / 5
Workflow Clarity	Sequencing is clear and includes explicit guards — the WRONG-vs-CORRECT create_agent pattern and the 'ensure that the APIs actually exist in the SDK' check act as validation checkpoints; post-edit testing is deferred to the run-locally skill rather than an inline verify step, leaving a minor gap.	4 / 5
Progressive Disclosure	Sections are well-organized with clear headers and outward pointers to sibling skills and external docs, but no bundle files exist so the per-class SDK API reference (ChatDatabricks, DatabricksEmbeddings, DatabricksVectorSearch, MCP variants) is inlined into SKILL.md rather than split into separate reference files.	3 / 5
	Total	15 / 20 Passed

Description

92%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is strong: it uses third-person imperative voice, names concrete actions, and pairs them with explicit, naturally-worded trigger phrases covering four distinct scenarios. It clearly distinguishes the skill from neighboring agent skills. Minor room remains to add a few synonyms or file extensions to push trigger-term coverage to fully comprehensive.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Modify agent code, add tools, or change configuration' plus 'Adding MCP servers', 'Changing agent instructions', 'Understanding SDK patterns' — giving comprehensive coverage of the agent-modification domain, matching the score-5 anchor.	5 / 5
Completeness	Explicitly states what ('Modify agent code, add tools, or change configuration') and when ('Use when: (1)...(4)...') with concrete trigger phrases, satisfying the score-5 anchor for answering both what and when.	5 / 5
Trigger Term Quality	Quotes natural phrases users would actually say verbatim ('modify agent', 'add tool', 'change model', 'edit agent.py'), which is strong keyword coverage; a few synonyms or file-extension variants are missing, so it sits just below the comprehensive score-5 anchor.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (editing the Databricks/langchain agent's code and config) with triggers tied to specific files like 'edit agent.py', minimizing overlap with sibling skills (add-tools, discover-tools, deploy).	5 / 5
	Total	19 / 20 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: databricks/app-templates
Path: agent-migration-from-model-serving/.claude/skills/modify-langgraph-agent/SKILL.md
Commit: fdc1b49

Reviewed: 1 day ago

Table of Contents

Discovery Implementation Validation

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.