update README CONCEPTS and SKILL COLLECTIONS for v2.1.121 + late-April research
CONCEPTS table: - Subagents: forked subagents on external builds via CLAUDE_CODE_FORK_SUBAGENT=1 (v2.1.117) - Hooks: PostToolUse can replace tool output via hookSpecificOutput.updatedToolOutput (v2.1.121) - MCP Servers: alwaysLoad bypasses tool-search deferral (v2.1.121) - Settings: add Auto Mode Config inline link - Ultrareview: claude ultrareview [target] non-interactive subcommand with --json/--timeout (v2.1.120) SKILL COLLECTIONS table: - mattpocock/skills star bump 33k → 36k, skill count 17 → 18 - wshobson/agents star drop 35k → 34k - rank swap: mattpocock/skills now ranks above wshobson/agents - VoltAgent/awesome-agent-skills curated count corrected 1,100+ → 930+ (badge overstates) Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
Shayan Rais
@@ -21,14 +21,14 @@ from vibe coding to agentic engineering - practice makes claude perfect
|
||||
|
||||
| Feature | Location | Description |
|
||||
|---------|----------|-------------|
|
||||
| <img src="!/tags/a.svg" height="14"> [**Subagents**](https://code.claude.com/docs/en/sub-agents) | `.claude/agents/<name>.md` | [](best-practice/claude-subagents.md) [](implementation/claude-subagents-implementation.md) Autonomous actor in fresh isolated context — custom tools, permissions, model, memory, and persistent identity |
|
||||
| <img src="!/tags/a.svg" height="14"> [**Subagents**](https://code.claude.com/docs/en/sub-agents) | `.claude/agents/<name>.md` | [](best-practice/claude-subagents.md) [](implementation/claude-subagents-implementation.md) Autonomous actor in fresh isolated context — custom tools, permissions, model, memory, and persistent identity · forked subagents available on external builds via `CLAUDE_CODE_FORK_SUBAGENT=1` (v2.1.117+) |
|
||||
| <img src="!/tags/c.svg" height="14"> [**Commands**](https://code.claude.com/docs/en/slash-commands) | `.claude/commands/<name>.md` | [](best-practice/claude-commands.md) [](implementation/claude-commands-implementation.md) Knowledge injected into existing context — simple user-invoked prompt templates for workflow orchestration |
|
||||
| <img src="!/tags/s.svg" height="14"> [**Skills**](https://code.claude.com/docs/en/skills) | `.claude/skills/<name>/SKILL.md` | [](best-practice/claude-skills.md) [](implementation/claude-skills-implementation.md) Knowledge injected into existing context — configurable, preloadable, auto-discoverable, with context forking and progressive disclosure · [Official Skills](https://github.com/anthropics/skills/tree/main/skills) · [Skills for Mono-repos](reports/claude-skills-for-larger-mono-repos.md) |
|
||||
| [**Workflows**](https://code.claude.com/docs/en/common-workflows) | [`.claude/commands/weather-orchestrator.md`](.claude/commands/weather-orchestrator.md) | [](orchestration-workflow/orchestration-workflow.md) Step-by-step guides for exploring codebases, fixing bugs, refactoring, and testing — orchestration patterns for multi-step tasks |
|
||||
| [**Hooks**](https://code.claude.com/docs/en/hooks) | `.claude/hooks/` | [](https://github.com/shanraisshan/claude-code-hooks) [](https://github.com/shanraisshan/claude-code-hooks) User-defined handlers (scripts, HTTP, MCP tools, prompts, agents) that run outside the agentic loop on specific events · [Guide](https://code.claude.com/docs/en/hooks-guide) |
|
||||
| [**MCP Servers**](https://code.claude.com/docs/en/mcp) | `.claude/settings.json`, `.mcp.json` | [](best-practice/claude-mcp.md) [](.mcp.json) Model Context Protocol connections to external tools, databases, and APIs |
|
||||
| [**Hooks**](https://code.claude.com/docs/en/hooks) | `.claude/hooks/` | [](https://github.com/shanraisshan/claude-code-hooks) [](https://github.com/shanraisshan/claude-code-hooks) User-defined handlers (scripts, HTTP, MCP tools, prompts, agents) that run outside the agentic loop on specific events — PostToolUse hooks can replace tool output via `hookSpecificOutput.updatedToolOutput` (v2.1.121+) · [Guide](https://code.claude.com/docs/en/hooks-guide) |
|
||||
| [**MCP Servers**](https://code.claude.com/docs/en/mcp) | `.claude/settings.json`, `.mcp.json` | [](best-practice/claude-mcp.md) [](.mcp.json) Model Context Protocol connections to external tools, databases, and APIs — `alwaysLoad` bypasses tool-search deferral so a server's tools are always loaded into context (v2.1.121+) |
|
||||
| [**Plugins**](https://code.claude.com/docs/en/plugins) | distributable packages | Bundles of skills, subagents, hooks, MCP servers, and LSP servers · [Marketplaces](https://code.claude.com/docs/en/discover-plugins) · [Create Marketplaces](https://code.claude.com/docs/en/plugin-marketplaces) |
|
||||
| [**Settings**](https://code.claude.com/docs/en/settings) | `.claude/settings.json` | [](best-practice/claude-settings.md) [](.claude/settings.json) Hierarchical configuration system · [Permissions](https://code.claude.com/docs/en/permissions) · [Model Config](https://code.claude.com/docs/en/model-config) · [Output Styles](https://code.claude.com/docs/en/output-styles) · [Sandboxing](https://code.claude.com/docs/en/sandboxing) · [Keybindings](https://code.claude.com/docs/en/keybindings) |
|
||||
| [**Settings**](https://code.claude.com/docs/en/settings) | `.claude/settings.json` | [](best-practice/claude-settings.md) [](.claude/settings.json) Hierarchical configuration system · [Permissions](https://code.claude.com/docs/en/permissions) · [Model Config](https://code.claude.com/docs/en/model-config) · [Output Styles](https://code.claude.com/docs/en/output-styles) · [Sandboxing](https://code.claude.com/docs/en/sandboxing) · [Keybindings](https://code.claude.com/docs/en/keybindings) · [Auto Mode Config](https://code.claude.com/docs/en/auto-mode-config) |
|
||||
| [**Status Line**](https://code.claude.com/docs/en/statusline) | `.claude/settings.json` | [](https://github.com/shanraisshan/claude-code-status-line) [](.claude/settings.json) Customizable status bar showing context usage, model, cost, and session info |
|
||||
| [**Memory**](https://code.claude.com/docs/en/memory) | `CLAUDE.md`, `.claude/rules/`, `~/.claude/rules/`, `~/.claude/projects/<project>/memory/` | [](best-practice/claude-memory.md) [](CLAUDE.md) Persistent context via CLAUDE.md files and `@path` imports · [Auto Memory](https://code.claude.com/docs/en/memory) · [Auto Memory Deep-dive](reports/claude-agent-memory.md) · [Rules](https://code.claude.com/docs/en/memory#organize-rules-with-clauderules) |
|
||||
| [**Checkpointing**](https://code.claude.com/docs/en/checkpointing) | automatic (git-based) | Automatic tracking of file edits with rewind (`Esc Esc` or `/rewind`) and targeted summarization |
|
||||
@@ -41,7 +41,7 @@ from vibe coding to agentic engineering - practice makes claude perfect
|
||||
| Feature | Location | Description |
|
||||
|---------|----------|-------------|
|
||||
| [**Routines**](https://code.claude.com/docs/en/routines)  | `claude.ai/code/routines`, `/schedule` | Cloud automation on Anthropic infrastructure — scheduled, API-triggered, or GitHub event-driven tasks that run even when your machine is off · [Desktop Tasks](https://code.claude.com/docs/en/desktop-scheduled-tasks) |
|
||||
| [**Ultrareview**](https://code.claude.com/docs/en/ultrareview)  | `/ultrareview` | Cloud-based multi-agent code review — every finding independently reproduced and verified in a remote sandbox (5–10 min, ~$5–20/run); 3 free runs for Pro/Max · [Tasks tracking](https://code.claude.com/docs/en/ultrareview#track-a-running-review) |
|
||||
| [**Ultrareview**](https://code.claude.com/docs/en/ultrareview)  | `/ultrareview`, `claude ultrareview [target]` | Cloud-based multi-agent code review — every finding independently reproduced and verified in a remote sandbox (5–10 min, ~$5–20/run); 3 free runs for Pro/Max · `claude ultrareview [target]` non-interactive subcommand for CI with `--json` and `--timeout` flags (v2.1.120+) · [Tasks tracking](https://code.claude.com/docs/en/ultrareview#track-a-running-review) |
|
||||
| [**Devcontainers**](https://code.claude.com/docs/en/devcontainer) | `.devcontainer/` | Preconfigured development containers with security isolation and firewall rules for consistent Claude Code environments |
|
||||
| [**Channels**](https://code.claude.com/docs/en/channels)  | `--channels`, plugin-based | Push events from Telegram, Discord, or webhooks into a running session — Claude reacts while you're away · [Reference](https://code.claude.com/docs/en/channels-reference) |
|
||||
| [**Ultraplan**](https://code.claude.com/docs/en/ultraplan)  | `/ultraplan` | Draft plans in the cloud with browser-based review, inline comments, and flexible execution — remotely or teleported back to terminal |
|
||||
@@ -135,10 +135,10 @@ Repos primarily known as curated libraries of `SKILL.md` files (distinct from fu
|
||||
| Name | ★ | <img src="!/tags/s.svg" height="14"> |
|
||||
|------|---|---|
|
||||
| [anthropics/skills](https://github.com/anthropics/skills) | 125k | 17 |
|
||||
| [wshobson/agents](https://github.com/wshobson/agents) | 35k | 152 |
|
||||
| [mattpocock/skills](https://github.com/mattpocock/skills) | 33k | 17 |
|
||||
| [mattpocock/skills](https://github.com/mattpocock/skills) | 36k | 18 |
|
||||
| [wshobson/agents](https://github.com/wshobson/agents) | 34k | 152 |
|
||||
| [scientific-agent-skills](https://github.com/K-Dense-AI/scientific-agent-skills) | 20k | 134 |
|
||||
| [awesome-agent-skills](https://github.com/VoltAgent/awesome-agent-skills) | 19k | 1,100+ (curated list) |
|
||||
| [awesome-agent-skills](https://github.com/VoltAgent/awesome-agent-skills) | 19k | 930+ (curated list) |
|
||||
|
||||
<p align="center">
|
||||
<img src="!/claude-jumping.svg" alt="section divider" width="60" height="50">
|
||||
|
||||
Reference in New Issue
Block a user