Skip to main content
Glama

install_skill

Install one reviewed SKILL.md across multiple agent targets at once; it runs as a dry run unless apply is confirmed.

Instructions

Install a SKILL.md for several agents at once (dry run unless apply=true).

skill_md: the reviewed SKILL.md text (or pass candidate_id to install the unedited draft). targets: "claude" (/.claude/skills) and/or "agents" (/.agents/skills, read by Codex and other Agent Skills clients), or "codex" (~/.codex/skills, older Codex). scope: "user" or "project". Existing skills are never replaced unless overwrite=true.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "install_skill"
  }
}

Input Schema

NameRequiredDescriptionDefault
applyNo
scopeNouser
projectNo
targetsNo
skill_mdNo
overwriteNo
candidate_idNo

Output Schema

NameRequiredDescriptionDefault
resultYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.4.0

TDQS

A4.4/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

The only annotation is destructiveHint=false, so the description carries most of the burden and does so well: it discloses the dry-run default (apply=true to commit), the exact filesystem destinations per target, and the non-destructive guard (no replacement unless overwrite=true). It does not describe what the dry run returns or whether installation is atomic across multiple targets, which keeps it from a 5.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loads the dry-run default in the first sentence, then uses a compact per-parameter glossary. Every clause carries information an agent needs — no restated title, no filler, no redundant safety boilerplate.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 7-parameter mutation tool with an output schema (so return values need no explanation), this is nearly complete: defaults, destinations, and safeguards are all present. The gaps are the undocumented `project` parameter and the unstated interaction between scope="project" and project.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 0%, so the description must compensate, and it explains 6 of 7 parameters — apply, skill_md, candidate_id, targets (with enumerated values), scope ("user" or "project"), and overwrite, including the meaning of each default. It leaves `project` entirely unexplained and never states that skill_md/candidate_id is effectively required despite the schema declaring no required parameters.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Uses a specific verb+resource ("Install a SKILL.md") and immediately scopes it: multiple agents at once, dry-run by default. The `candidate_id` clause implicitly distinguishes installing a reviewed file from installing an unedited draft, so an agent can tell what this does without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives concrete selection guidance among the tool's own modes: pass `skill_md` for reviewed text vs `candidate_id` for the unedited draft, and pick a `targets` value based on the client (claude/agents/codex). It does not say when to prefer this over siblings such as draft_skill, and has no explicit when-not guidance, so it stops short of a 5.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.