Reserved name placeholder. No functionality.

Command-line client for the Jentrix MCP surface (jentrix tool <name>, generated noun-verb commands).

> Claude Code compatibility layer: trust model, file commands, hooks, MCP servers, and tool name mapping.

OpenTelemetry traces for pi-coding-agent — per-prompt span tree (interaction → llm_request, tool.<name>) exported via OTLP. Aspire-dashboard ready.

Personal Headroom compression bridge for Pi with Docker-first defaults and Pi tool-name adaptation

Declarative tool-call policy for DeepSeek Harness: config-driven allow/deny/ask rules (tool name, args pattern, shell command prefix/regex, priority) on the tools/pre-execute gate — Gemini CLI-style policy files expressed as plain plugin config.

Anthropic Tool Name Proxy — translates tool names > 64 chars for OpenAI-compatible endpoints

Declarative Claude Code-style permission rules plus a Codex-style process-level network policy for DeepSeek Harness: ordered allow/deny/ask rules with tool-name, argument (glob/regex), workspace-path, and network-target (domain/ip/port/scheme) matching on

MCP security scanner: scan MCP servers and tool manifests for tool poisoning, prompt injection, tool-name shadowing, excessive-agency schemas, unpinned/remote sources, and rug-pulls - 10 OWASP-mapped heuristics, CLI + MCP server, fully offline. Also runs

Stable, collision-safe tool names across MCP, OpenAI, Anthropic, and Gemini.

See what your TypeScript AI agent actually did. Wrap a tool in one line; every call is recorded locally (tool name, argument names, never values) and `audit` prints the report. Nothing is blocked.

Transparent proxy that adds/strips tool name prefixes for Claude API requests

DSH 工具卡片标题中文化：Code/Read/Bash 等工具名显示为中文。

OpenCode plugin for using AnyRouter as a Claude API provider, fixes tool name casing and array serialization issues

Detect if code is running in an AI agent or automated development environment

Cline-style plan and act workflow for OpenCode - dynamically fetches latest prompts from Cline's official repository with smart caching and automatic tool name mapping

Tool-call timeout policy: a tools/execute wrapper that arms a per-tool deadline on exec.signal and returns TOOL_TIMEOUT when it wins

Model-facing skill loading tool for the DeepSeek Harness

Repeat-tool-call guard plugin: advisory reminders when an agent loops on identical tool calls

An object-oriented command-line parser for TypeScript