Loading...
Loading...
Found 4,998 Skills
GitLab label operations. ALWAYS use this skill when user wants to: (1) list project labels, (2) create new labels, (3) manage label colors and descriptions.
Create a git commit following project conventions
GitLab repository operations. ALWAYS use this skill when user wants to: (1) clone repositories, (2) fork projects, (3) view repo info, (4) create new projects, (5) archive/delete repos, (6) manage repo settings.
GitLab search operations via API. ALWAYS use this skill when user wants to: (1) search across GitLab globally, (2) find issues/MRs/code/commits, (3) search within a group or project, (4) find users or projects by keyword.
GitLab protected branch operations via API. ALWAYS use this skill when user wants to: (1) view branch protection rules, (2) protect/unprotect branches, (3) configure push/merge access levels, (4) set up code owner approval requirements.
This skill is used when users explicitly request "review NSFC proposals", "simulate expert review", or "evaluate NSFC applications". It simulates the perspective of domain experts to conduct multi-dimensional reviews of NSFC proposals, outputting graded issues and actionable modification suggestions. ⚠️ Not applicable: when users only want to write/modify a specific section of a proposal (use the nsfc-*-writer series skills instead), only want to understand review criteria (answer directly), or have no clear "review/evaluate" intent.
Systematically diagnose and resolve bugs through conversational investigation and root cause analysis
Generate a personalized follow-up sequence for any creator chasing scenario — missing info, unsigned contract, late content, missing metrics, or incomplete whitelisting setup. This skill should be used when chasing a creator for a response, writing a follow-up message to an influencer, nudging a creator about a late deliverable, following up on an unsigned contract, requesting missing campaign metrics, chasing whitelisting or ad access setup, escalating a non-responsive creator, writing a reminder to a creator who ghosted, building a follow-up cadence for overdue items, drafting a polite but firm nudge to an influencer, or managing creator communication when deadlines slip. For writing initial outreach messages, see creator-outreach-sequence-generator. For classifying and triaging creator replies, see reply-triage-classifier. For negotiating rates after a creator responds, see creator-negotiation-assistant.
Structured checkpoint format for requesting human input. When an agent needs a decision, it must stop, present context, show options, and wait. Activate when delegating to subagents, running background tasks, or hitting any decision point that requires human judgment.
Use this skill whenever the user asks to analyze, understand, or survey an entire project, codebase, or any collection of files. Trigger phrases include "analyze a large file", "process multiple files", "comprehend this problem", "take a look at these files", "familiarize yourself with this project", or any similar request, however phrased. Also activate when the task involves processing context that exceeds what can be reasoned about in a single pass, when encountering any input larger than ~50KB that requires detailed analysis, or when the user mentions "context comprehension" or "recursive comprehension". This skill TAKES PRIORITY over your default explore subagents for any project-wide or codebase-wide analysis task.
Use when the user wants to commit changes. Stages files, updates CHANGELOG.md, and creates a commit following project conventions.
File-based knowledge persistence patterns: when to store discoveries, when to recall past solutions, and how to organize project memory. Activate when starting tasks, encountering errors, making decisions, or when context may be lost between sessions.