Skip to main content
Glama
Sign In

Oxyo (oxyo.si) — Cut Coding Agent Token & Quota Burn by up to 72%

by io.github.Eno654
OverviewDiscussionsAdmin

oxyo_sleep_abstract

Idempotent

Distill verified code execution traces into tested, reusable functions in the persistent skill vault while compacting older episodic logs. Use after coding milestones to extract skills; use oxyo_consolidate for episode cleanup without skill synthesis.

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": "oxyo_sleep_abstract"
  }
}

Input Schema

NameRequiredDescriptionDefault
scopeNoOptional project/repo scope to process.
tracesNoOptional verified trace objects [{'prompt': ..., 'verified_code': ...}] to abstract.
keep_recentNoNumber of newest episodes retained verbatim during memory compaction.
max_new_skillsNoMaximum number of newly synthesized skills promoted in one pass (1..10).

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed4 schema fields changed
    • addedInput schema / properties / keep_recent / description
      Added value: +"Number of newest episodes retained verbatim during memory compaction."
    • addedInput schema / properties / max_new_skills / description
      Added value: +"Maximum number of newly synthesized skills promoted in one pass (1..10)."
    • addedInput schema / properties / scope / description
      Added value: +"Optional project/repo scope to process."
    • changedInput schema / properties / traces / description
      Previous value: -"Optional inline verified trace objects [{'prompt': ..., 'verified_code': ...}]"New value: +"Optional verified trace objects [{'prompt': ..., 'verified_code': ...}] to abstract."
  2. First observed

TDQS

A4.4/5.0
Behavior4/5

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

Annotations only declare it is a non-read-only, idempotent, non-destructive write; the description adds that output lands in a persistent skill vault and that older episodic logs are compacted as a side effect, which is behavior an agent could not infer from the flags. It does not say how traces are selected when the `traces` argument is omitted, nor whether compaction is reversible — a notable gap since `destructiveHint: false` sits slightly in tension with the lossy-sounding "compacting" wording.

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 sentences with zero filler: the core operation and its side effect come first, then the routing guidance. Every clause ("tested", "verified", "persistent") carries distinct information.

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 4-parameter, all-optional tool with a fully documented schema and no output schema, the description covers purpose, trigger, and the main competing tool. What it leaves out is the no-argument default path (does it mine recent episodes when `traces` is absent?) and what the promotion of skills looks like on return — modest gaps rather than blockers.

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 description coverage is 100%, so all four parameters already carry their own documentation and the baseline applies. The description does clarify that only *verified* traces are eligible for abstraction, which sharpens the meaning of the `traces` argument, but adds nothing about `scope`, `keep_recent`, or `max_new_skills` beyond 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 (Distill) over a specific input (verified code execution traces) producing a specific artifact (tested, reusable functions in the persistent skill vault), plus a secondary effect (compacting older episodic logs). It also names the sibling it is not (`oxyo_consolidate`), so an agent can distinguish the two without opening a schema.

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

Usage Guidelines5/5

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

Gives an explicit trigger ("after coding milestones") and an explicit alternative with its selection condition ("use `oxyo_consolidate` for episode cleanup without skill synthesis"). The when-to-use / when-to-use-something-else pair is complete for this decision.

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

All Oxyo (oxyo.si) — Cut Coding Agent Token & Quota Burn by up to 72% toolsInspect
Try in Browser
Claude Code, inside the Godot 4 editorFlockbay puts an AI agent in the real Godot 4 editor: it writes GDScript, runs your game and fixes what breaks. Free, or bring your own Claude Code or Codex.flockbay.comAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources