@cablate/agentic-mcp
BAgentic MCP - Three-layer progressive disclosure for MCP servers with Socket daemon. Use when the user needs to interact with MCP servers, query available tools, call MCP tools, or manage the MCP daemon process. Provides socket-based communication for efficient server interaction with three-layer progressive disclosure API.
Install
agr install @cablate/agentic-mcp --target claudeWrites 11 files into .claude/skills/, pinned to git-557b71f0.
- .claude/skills/agentic-mcp/.gitignore
- .claude/skills/agentic-mcp/Banner.png
- .claude/skills/agentic-mcp/CHANGELOG.md
- .claude/skills/agentic-mcp/LICENSE
- .claude/skills/agentic-mcp/README.md
- .claude/skills/agentic-mcp/README_zhTW.md
- .claude/skills/agentic-mcp/SKILL.md
- .claude/skills/agentic-mcp/package-lock.json
- .claude/skills/agentic-mcp/package.json
- .claude/skills/agentic-mcp/tsconfig.json
- .claude/skills/agentic-mcp/vitest.config.ts
Document
name: agentic-mcp description: Agentic MCP - Three-layer progressive disclosure for MCP servers with Socket daemon. Use when the user needs to interact with MCP servers, query available tools, call MCP tools, or manage the MCP daemon process. Provides socket-based communication for efficient server interaction with three-layer progressive disclosure API.
Agentic MCP
Quick start
agentic-mcp daemon start --config <mcp-servers.json path> # Start daemon
agentic-mcp metadata <server-name> # Get server info
agentic-mcp list <server-name> # List tools
agentic-mcp schema <server-name> <tool-name> # Get tool schema
agentic-mcp call <server-name> <tool-name> --params '{"arg":"value"}' # Call tool
agentic-mcp daemon stop # Stop daemon
Core workflow
agentic-mcp daemon start- Start daemonagentic-mcp metadata <server>- Layer 1: server infoagentic-mcp list <server>- Layer 2: available toolsagentic-mcp schema <server> <tool>- Layer 3: tool detailsagentic-mcp call <server> <tool> --params '{"arg":"value"}'- Execute tool
Daemon management
agentic-mcp daemon start # Start daemon
agentic-mcp daemon health # Check status
agentic-mcp daemon reload # Reload config
agentic-mcp daemon stop # Stop daemon
agentic-mcp daemon start --config <path> # Custom config
Sessions (isolated instances):
MCP_DAEMON_SESSION=<name> agentic-mcp daemon start
Query commands
agentic-mcp metadata <server> # Server info
agentic-mcp list <server> # List tools
agentic-mcp schema <server> <tool> # Tool schema
agentic-mcp daemon health # Check daemon status
Tool calls
agentic-mcp call <server> <tool> --params '{"argName":"value"}'
All parameters must be JSON object via --params.
JSON mode:
agentic-mcp metadata <server> --json
agentic-mcp call <server> <tool> --params '{"arg":"value"}' --json
Configuration
mcp-servers.json:
{
"servers": {
"<server-name>": {
"command": "npx",
"args": ["-y", "@scope/mcp-server"]
}
}
}
Socket protocol
Command (newline-delimited JSON):
{"id":"1","action":"metadata","server":"<server>"}
Response:
{"id":"1","success":true,"data":{...}}
Platform: Windows (TCP) / Unix (domain socket)
Examples
Basic usage:
agentic-mcp daemon start
agentic-mcp metadata <server>
agentic-mcp list <server>
agentic-mcp schema <server> <tool>
agentic-mcp call <server> <tool> --params '{"arg":"value"}'
Multiple sessions:
MCP_DAEMON_SESSION=proj1 agentic-mcp daemon start
MCP_DAEMON_SESSION=proj2 agentic-mcp daemon start
Debugging
agentic-mcp daemon health # Check status
agentic-mcp daemon reload # Reload after config change
Errors
✗ MCP daemon is not running
✗ Server '<name>' not found
✗ Tool '<name>' not found
✗ Required argument '<name>' not provided
Trustgrade B
- passBody integrity
Whether the stored document is plausibly the kind of file the artifact declares, rather than something fetched by mistake.
- passType matchnot applicable to this artifact type
Whether the artifact is really the kind of thing its metadata claims it is.
- passFreshness
How long since the source repository was last pushed to.
- passPrompt injection
Scans the artifact's own text for instructions aimed at your agent rather than at you.
- warnLicenseno SPDX license detected
Whether the source repository declares an SPDX license permissive enough to redistribute.
How the grade is calculated
Each check contributes 0 points when it passes, 1 when it warns, and 2 when it fails. The total maps to a letter:
- Aevery check passed
- Bone warning
- Ctwo warnings
- Dprompt injection or body integrity failed, or three warnings
- Fone of those failed, and something else is wrong
These are automated hygiene checks, not a security audit, and not a dependency or vulnerability scan. A grade of A means nothing was flagged — not that the artifact is safe.
Versions
git-557b71f00e832026-07-31