45ck/noslop/context-loader
Claude Code and agent skill by 45ck/noslop
- community
Checking that the SKILL.md file is still in the repository.
Load relevant codebase context based on task type. Trigger this skill when starting any task that involves understanding or modifying the Beacon codebase, especially when working in an unfamiliar area.
Install
npx vskill@latest install 45ck/noslop/context-loaderSKILL.md
Context Loader Skill
You help the agent load the right files for the current task instead of reading the entire codebase.
Instructions
When invoked with a task description, follow these steps:
Step 1: Read the Module Map
Read docs/module-map.md to understand the high-level codebase structure. Identify which modules are relevant to the described task.
Step 2: Match Against Context Selectors
Read docs/context-selectors.md and find the task type that best matches the description. If the task spans multiple categories, combine the selectors.
Task type mapping hints:
- Mentions "event", "tracking", "ingestion" → Adding a New Event Type
- Mentions "chart", "graph", "visualization" → Adding a New Chart Type
- Mentions "dashboard", "UI", "component" → Fixing a Dashboard Bug
- Mentions "login", "permission", "API key", "token" → Changing Auth Flow
- Mentions "billing", "subscription", "plan", "invoice" → Billing Issue
- Mentions "slow", "timeout", "latency", "memory" → Performance Problem
- Mentions "job", "worker", "queue", "scheduled" → Worker / Job Issue
- Mentions "schema", "API", "GraphQL", "endpoint" → API Schema Change
Step 3: Load Targeted Context
Read only the files listed in the matched selector's Read column. Do not read files outside this set unless you discover a dependency during the investigation.
Step 4: Summarize What You Loaded
Before starting work, output a brief summary:
Step 5: Check Layer-Specific Rules
If the relevant modules have their own CLAUDE.md files (e.g., src/query/CLAUDE.md), read those for module-specific constraints and patterns before writing code.
Notes
- If no selector matches, fall back to reading the module map summaries for
the two most relevant modules and explore from there.
- Keep token budget in mind. Prefer reading key files (validators, services,
resolvers) over utility or test files during the context-loading phase.
- If the task requires cross-cutting changes (more than 3 modules), flag
this to the user before proceeding.
Compaction
If total loaded context exceeds ~8,000 tokens:
- Identify entries with
last_verifiedolder than 60 days — summarize to one line each - Collapse implementation details, keep only interface signatures
- Remove examples, keep only the rule they demonstrate
- Target: reduce to ~5,000 tokens while preserving all rules and current-task context
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingexamples/long-horizon/06-context-compression/.claude/skills/context-loader/SKILL.md
Popularity
- Repository stars
- 14
- Repository forks
- 1
- 7-day trend score
- 19
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/45ck/noslop/context-loader)