Skip to main content
Glama

run_nav_generate

Generates Left 4 Dead 2 nav meshes by launching the game as a listen server, copying the map loose, hiding conflicting add-ons, and returning the .nav, nav dump, verdict, and warnings.

Instructions

Generates nav meshes the only way Left 4 Dead 2 can: it launches the game client as a listen server, with sv_cheats 1 and nav_edit 1, and waits for the engine to write each map's .nav. The map must be loose, so the tool copies it into left4dead2/maps, moves the named add-on archives aside so they cannot shadow it, and restores everything afterwards. Returns the mesh, the engine's <map>_nav.txt dump, the decoded nav verdict and the flow warnings from the console. A map whose archive contains the same map must be passed in addonsToHide. Long-running and guarded.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "maps": "<maps>"
    },
    "name": "run_nav_generate"
  }
}

Input Schema

NameRequiredDescriptionDefault
mapsYesCompiled .bsp files to generate nav for, absolute or relative to the repo.
outDirNoDirectory for the generated .nav and <map>_nav.txt. Defaults to the directory of each source .bsp.
confirmNo
addonsToHideNoAdd-on archives to move aside while generating (for example a campaign VPK that contains the same map names). Restored even if generation fails.
timeoutSecondsNoPer map.

Output Schema

NameRequiredDescriptionDefault
okYes
mapsYes
restoredYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.0.24-alpha

TDQS

A4.5/5.0
Behavior5/5

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

With no meaningful annotations beyond a title, the description carries the full burden and does so: it discloses the external process launch, that the map is copied into left4dead2/maps, that add-on archives are moved aside and restored even on failure, that it is long-running, and what the return payload contains. This is far beyond what structured fields provide.

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

Conciseness4/5

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

Four sentences, front-loaded with the core action and mechanism, then prerequisites and returns. The copy/hide/restore sentence is dense but every clause conveys a distinct side effect, so little is wasted.

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 complex, side-effecting, long-running tool with an output schema, it covers mechanism, prerequisites, filesystem side effects and restoration, and return contents. The only gap is the unexplained confirm gating, which an agent might need to know to avoid a failed or blocked call.

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 coverage is 80%, so baseline is 3; the description adds real meaning by explaining why addonsToHide exists (archives shadowing a same-named map) and by framing timeoutSeconds as per-map for a long-running operation. The confirm parameter and outDir defaults are left to the schema.

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?

States a specific verb and resource ('Generates nav meshes') and immediately explains the mechanism (launching the L4D2 client as a listen server with sv_cheats/nav_edit) so it is clearly distinct from read-only siblings like read_nav or read_nav_dump. An agent knows exactly what this tool produces.

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 prerequisites: the map must be loose, and any add-on archive containing the same map must be listed in addonsToHide. It does not, however, explicitly contrast itself with read_nav/read_nav_dump or explain what 'guarded' (the confirm flag) means for when the call will actually proceed.

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