Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
/plugin marketplace add knitli/toolshednpx agentmods add plugins/knitli/toolshed/codeweavergit clone --depth 1 https://github.com/knitli/toolshedGrade A, and why
codeweaver scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
{
"userConfig": {
"VOYAGE_API_KEY": {
"type": "string",
"title": "Voyage AI API Key",
"description": "API key for Voyage AI embedding provider (get one at https://voyage.ai)",
"sensitive": true,
"required": false
}
},
"name": "codeweaver",
"version": "1.0.1",
"description": "Exquisite Context for AI Agents — Semantic code search MCP server with hybrid search, AST-level understanding, and intelligent chunking for 166+ languages.",
"author": {
"name": "Knitli Inc.",
"email": "[email protected]"
},
"homepage": "https://knitli.com",
"repository": "https://github.com/knitli/toolshed",
"license": "MIT OR Apache-2.0",
"keywords": [
"mcp",
"search",
"semantic",
"code-search",
"semantic-search",
"AST",
"indexing",
"code-intelligence",
"embeddings",
"vector-search",
"hybrid-search",
"codebase-analysis"
],
"mcpServers": {
"codeweaver": {
"command": "uvx",
"args": [
"--from",
"code-weaver['recommended']",
"cw",
"server",
"--transport",
"stdio"
],
"env": {
"XDG_DATA_HOME": "${CLAUDE_PLUGIN_DATA}/data",
"XDG_CACHE_HOME": "${CLAUDE_PLUGIN_DATA}/cache",
"XDG_STATE_HOME": "${CLAUDE_PLUGIN_DATA}/state",
"XDG_CONFIG_HOME": "${CLAUDE_PLUGIN_DATA}/config",
"VOYAGE_API_KEY": "${user_config.VOYAGE_API_KEY}"
}
}
}
}
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 56 lines scan A 33d5ca2ed0ef
codeweaver is a plugin published in the GitHub repository knitli/toolshed (1 stars, last pushed 4mo ago), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other plugins, from other repositories
motherduck-skills
Plugin marketplace listing 1 plugin: motherduck-skills.
motherduck-skills
MotherDuck skills for agents and developers: use the CLI and Guides, explore schemas, write DuckDB SQL, build Dives and Flights, and ship analytics apps and pipelines.
frappe-agent
Plugin marketplace listing 1 plugin: frappe-agent.
frappe-agent
Frappe, ERPNext, and Frappe ecosystem coding, customization, bench, and review intelligence for Claude Code and AI coding workflows.
supergraph
Evidence-gated AI coding workflow: scan → analyze → plan → TDD → execute → fix → verify → review, powered by Codebase Memory MCP >= 0.9.0 with optional Serena LSP intelligence. Includes blast-radius planning, test/cycle gates, independent review, and Windows Git Bash hook auto-resolution.
supergraph
Plugin marketplace listing 1 plugin: supergraph.