cubrid-manual
Skill0 starsChanged 14 days agoMITView on GitHub
Look up CUBRID engine behavior in the official version-pinned manual and return a citable URL: SQL syntax, functions, data types, reserved words, config parameters, driver APIs. Use it before writing dialect or driver code, or when an answer about CUBRID behavior needs a source rather than memory. For CMT questions, use cmt-manual. Triggers on phrases like 'CUBRID 매뉴얼 확인', 'CUBRID에서 X는 어떻게 동작해', 'is X reserved in CUBRID'.
In this repository22 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- blogSkill
- ci-checkSkill
- cmt-manualSkill
- commitSkill
- hhh-dialect-verify-reportSkill
- hhh-dialect-verifySkill
- java-formatSkill
- jdbc-ctp-verify-reportSkill
- jira-draftSkill
- jira-fetchSkill
- noteSkill
- pr-draftSkill
- pr-reviewSkill
- reportSkill
- review-triageSkill
- self-reviewSkill
- skill-createSkill
- submodule-bumpSkill
- tutorSkill
- weekly-reportSkill
- worklogSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 24
Default to English (en). Use Korean (ko) if the user writes in Korean or asks for the Korean manual.
- Distinctive: line 47
If you know the page path from Step 3, WebFetch it directly with a focused prompt.
- Distinctive: line 48
If unsure which page covers the topic, WebSearch first, then WebFetch the best hit:
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
CUBRID manual lookup (online)
Answer CUBRID engine questions from the official version-pinned online manual, and return a citable URL ready to paste into a PR description or verification report.
Step 1: Determine the version
CUBRID semantics differ by version, so never guess. Read the version from $ARGUMENTS (e.g. 10.2, 11.4). If none is given, ask the user: typically 10.2 (the dialect's minimum supported version) or 11.4 (current). Use it in every URL below.
Step 2: Determine the language
Default to English (en). Use Korean (ko) if the user writes in Korean or asks for the Korean manual. Only the en/ko segment of the URL changes.
Step 3: Map the topic to a manual page
sql/syntax.html·sql/datatype.html·sql/identifier.html·sql/keyword.html·sql/literal.html·sql/transaction.html·sql/partition.html: core SQL, types, identifiers, reserved wordssql/function/: built-in functions (string, numeric, datetime, JSON, aggregate;analysis_fn.htmlfor window/analytic)sql/query/: DML:select.html,insert.html,update.html,delete.html,merge.html,cte.htmlsql/schema/: DDL:table_stmt.html(CREATE TABLE), etc.api/: drivers:jdbc.html,cci.html, and othersadmin/: server admin, utilities, configuration parameterspl/: stored procedures (PL/CSQL, Java SP)- top-level:
csql.html,ha.html,shard.html,security.html,install.html
Step 4: Fetch the version-pinned page
Build the URL:
https://www.cubrid.org/manual/<lang>/<version>/<path>.html[#anchor]Example: https://www.cubrid.org/manual/en/10.2/sql/query/select.html#for-update.
- If you know the page path from Step 3, WebFetch it directly with a focused prompt.
If unsure which page covers the topic, WebSearch first, then WebFetch the best hit:
WebSearch query="<term> site:cubrid.org/manual/<lang>/<version>"
Step 5: Answer and cite
Give the documented answer concisely, then cite the version-pinned URL (with #anchor when applicable) so the result is ready to paste into a PR description or verification report. If the manual is silent or contradicts observed behavior, say so explicitly rather than guessing.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

