Skip to main content
Glama
io.github.sionjang101-create/taskattest

io.github.sionjang101-create/taskattest

Draft a spec from the files

draft_spec
Idempotent

Reads input, output, and trace files to propose a spec with output structure, input seals, ID links, trace ordering/counts, and field profiles; deterministic, flagged for human review.

Instructions

Read the input, output and trace files and propose a spec: structure checks for outputs, seals on the inputs, checks relating output ids to input ids, and for traces the ordering and counts it can infer. Also returns a profile of every field so you can write your own conditions without opening the files. Deterministic, no model calls. It never fits a threshold to an output; the spec is marked draft and every verdict against it carries a warning until a person removes the flag.

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": {
      "spec_path": "<spec_path>"
    },
    "name": "draft_spec"
  }
}

Input Schema

NameRequiredDescriptionDefault
writeNowrite the draft to spec_path; refuses to overwrite
inputsNoname -> path of files given to the agent
tracesNostate logs, read in order
outputsNoname -> path of files the agent must produce
task_idNo
spec_pathYesWhere the spec will live; paths are made relative to it. Written only if write is true.
descriptionNo
line_prefixNoonly lines starting with this are records; detected if omitted

Output Schema

NameRequiredDescriptionDefault
specYes
notesYes
profilesYes
written_toYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A3.9/5.0
Behavior5/5

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

Beyond the annotations, the description discloses that the operation is deterministic with no model calls, that it never fits a threshold to an output, that the spec is marked draft, and that every verdict carries a warning until a human clears the flag. The "refuses to overwrite" behavior is consistent with destructiveHint=false and readOnlyHint=false, and the warning/draft semantics are exactly the kind of consequence an agent could not derive from the 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.

Conciseness4/5

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

Three sentences, front-loaded with the core action, and each carries distinct information: what is produced, what side benefit comes back, and how the artifact is constrained. The middle sentence's em-dash clauses run a bit long but nothing is redundant.

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, the description does not need to explain return values, and the annotations cover safety. What it supplies — determinism, draft marking, human-in-the-loop clearing — is the right depth for an 8-parameter tool. The remaining gap is that no parameter-level hints or comparison to verify_task are offered.

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 75%, close to the threshold where the schema carries its own weight, and the description only loosely maps "input, output and trace files" onto the inputs/outputs/traces parameters. It adds no meaning for line_prefix, task_id, description, or the write flag's interaction with spec_path, so it neither compensates for nor meaningfully extends the schema.

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?

"Read the input, output and trace files and propose a spec" names a specific verb and resource, and the sentence enumerates exactly what the resulting spec contains (structure checks on outputs, seals on inputs, id->output relations, trace ordering/counts). It is unambiguous on its own, but it never references a sibling, so an agent cannot tell from the text alone whether this complements verify_task or describe_checks.

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?

Usage is implied rather than stated: the tool is meant to be run before verification exists, and the draft flag is removed later by a person. There is no explicit when-to-use, when-not-to-use, or named alternative, so an agent must infer its place in the workflow from the sibling list.

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