oracle
Best practices for using the oracle CLI (prompt + file bundling, engines, sessions, and file attachment patterns).
64
48%
Does it follow best practices?
Impact
99%
1.70xAverage score across 3 eval scenarios
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/oracle/SKILL.mdQuality
Discovery
32%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 identifies a specific tool (oracle CLI) and lists topic areas but lacks concrete actions and an explicit 'Use when...' clause. It reads more like a table of contents than a skill description, making it difficult for Claude to confidently select this skill over others in a large skill set.
Suggestions
Add an explicit 'Use when...' clause, e.g., 'Use when the user asks about the oracle CLI, oracle commands, or needs help with oracle prompt construction, file bundling, or session management.'
Replace 'Best practices for using' with specific concrete actions, e.g., 'Guides construction of oracle CLI prompts, configures engines, manages sessions, and attaches files using correct bundling patterns.'
Include natural trigger terms users might say, such as 'oracle command', 'oracle prompt', 'oracle session', or 'oracle file attachment' to improve discoverability.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain ('oracle CLI') and lists some areas (prompt + file bundling, engines, sessions, file attachment patterns), but these are topic areas rather than concrete actions. It says 'best practices for using' which is vague about what specific actions the skill enables. | 2 / 3 |
Completeness | It partially addresses 'what' (best practices for oracle CLI usage) but there is no explicit 'when' clause or trigger guidance. Per the rubric, a missing 'Use when...' clause caps completeness at 2, and the 'what' itself is also weak (just 'best practices'), so this scores a 1. | 1 / 3 |
Trigger Term Quality | Includes relevant keywords like 'oracle CLI', 'prompt', 'file bundling', 'engines', 'sessions', and 'file attachment patterns'. However, it's unclear if users would naturally say these terms—'oracle CLI' is specific but niche, and common variations or natural phrasings are missing. | 2 / 3 |
Distinctiveness Conflict Risk | The mention of 'oracle CLI' provides some distinctiveness, but 'prompt', 'file bundling', 'sessions', and 'file attachment patterns' are generic enough terms that could overlap with other CLI tools or file-handling skills. The niche is somewhat clear but not sharply defined. | 2 / 3 |
Total | 7 / 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 is a solid, practical skill that provides concrete CLI commands and patterns for using the oracle tool. Its main strengths are actionability (specific commands with real flags and glob patterns) and reasonable organization. Weaknesses include missing validation/error-recovery steps in the workflow and some verbosity in sections that explain concepts Claude would already understand (prompt writing advice, what 'one-shot' means).
Suggestions
Add explicit validation checkpoints to the golden path — e.g., 'If dry-run shows >X tokens, reduce file set' and 'After receiving output, verify claims against actual code before applying changes'.
Trim the prompt template section to just the bullet list of what to include, removing the preamble about oracle having 'zero project knowledge' — Claude can infer why these elements matter.
Consider adding a brief error-handling section: what to do when browser automation fails, when sessions are lost, or when token limits are exceeded.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Generally efficient but includes some unnecessary explanations (e.g., explaining what oracle does conceptually, the 'Oracle starts with zero project knowledge' paragraph restates what's obvious). The prompt template section borders on teaching Claude how to write good prompts, which it already knows. Some sections like the 'exhaustive prompt restoration pattern' could be tightened. | 2 / 3 |
Actionability | Provides concrete, copy-paste-ready commands throughout — dry-run, browser run, file attachment patterns with glob syntax, session reattachment, remote browser hosting. The file inclusion/exclusion examples and engine selection logic are specific and executable. | 3 / 3 |
Workflow Clarity | The 'golden path' provides a clear 4-step sequence, and the session reattachment flow is well-documented. However, there are no explicit validation checkpoints or error recovery loops — e.g., what to do if dry-run shows too many tokens, what to do if browser run fails (beyond timeout/reattach), or how to verify oracle output against code+tests as mentioned in the intro. | 2 / 3 |
Progressive Disclosure | Content is well-organized with clear section headers and logical grouping. However, with no bundle files, everything is inline in a single file. The prompt template section and exhaustive prompt pattern could potentially be split out. The file is ~100 lines which is manageable but some sections (like engine details, remote browser hosting) could be referenced rather than inline. | 2 / 3 |
Total | 9 / 12 Passed |
Validation
72%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 8 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
metadata_version | 'metadata.version' is missing | Warning |
metadata_field | 'metadata' should map string keys to string values | Warning |
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 8 / 11 Passed | |
- Repository
- deepgram/dglabs-deepclaw
- Commit
ec8d4f8
- 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.