Agents-Core

Overview Schema Source Related Servers Discussions Admin

Capture local inference context

local_context_create

Create an immutable, workspace-bound context from source files and shared instructions without starting a model. Returns one-time bearer tokens for access.

Instructions

Capture an immutable workspace-bound context without starting a model.

Returns owner/access bearer tokens once; keep them outside prompts and logs.

Input Schema

TableJSON Schema

Name	Required	Description
`ttl_seconds`	No	Context lifetime 30..3600 seconds, default 900, further capped by owner.
`max_requests`	No	Context model-call budget 1..100, default 10, capped by owner.
`source_paths`	No	At most 32 unique regular UTF-8 files relative to source_root_id; default empty. No absolute paths, traversal or symlinks.
`common_system`	Yes	Required UTF-8 common system instructions; empty allowed. Context total capped by owner, default 256 KiB.
`shared_context`	Yes	Required UTF-8 shared data. Shared text plus captured files must contain data.
`source_root_id`	No	Optional owner-configured root ID, null by default; required with source_paths. No raw root path override.
`artifact_manifest`	No	Optional pinned active artifact manifest. The broker hydrates its exact selected artifacts before generation, checks byte/token budgets and rechecks access/integrity before dispatch, publication and result reads.
`max_total_output_tokens`	No	Context aggregate output budget 1..65536, default 4096, capped by owner.

Output Schema

TableJSON Schema

Name	Required	Description	Default
No arguments

Schema Changelog

Changes observed during successful MCP inspections.

Addedv26.10.16Oct 16, 2026

TDQS

A3.8/5.0

Behavior4/5

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

Annotations declare readOnly=false, idempotent=false, destructive=false, but the description adds real context beyond them: the context is 'immutable', tokens are returned only once, and they must be kept outside prompts and logs. This one-time secret exposure is exactly the kind of behavioral detail an agent needs and cannot get from structured fields.

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?

Two tight sentences with zero waste. The core capability is front-loaded and the security constraint follows immediately; every clause earns its place.

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?

With an output schema present, full parameter coverage, and annotations covering the safety profile, the description only needs to carry the non-obvious behavioral bits, which it does (immutability, one-time token return). Minor gap: no explicit note on the create→grant→generate lifecycle position relative to siblings.

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

Parameters3/5

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

Schema coverage is 100%, so all eight parameters are already richly documented (ttl bounds, max_requests budget, path constraints, owner caps). The description adds no additional parameter meaning, so baseline 3 is appropriate.

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

Purpose4/5

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

States a specific verb (Capture) and resource (immutable workspace-bound context) and adds the differentiator 'without starting a model', which separates it from local_generate. It does not explicitly name the sibling alternatives, so it's clear but lacks sibling routing.

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

Usage Guidelines3/5

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

The phrase 'without starting a model' implies this is a preparatory step before generation, but there is no explicit when-to-use, when-not, or reference to siblings like local_context_grant or local_context_release. Usage must be inferred from the lifecycle.

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

Deploy Server