agentsclimarketplace

Inline code

Skill sangjinsu/skill-execution-bench/skills/inline-code

Local-first benchmark comparing four Skill packaging patterns — doc-only, inline-code, Python script, and Go binary — to find which is most reliable when an LLM coding agent executes a Skill.From the repository description

Install
npx -y skills add sangjinsu/skill-execution-bench --skill inline-code

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

3.1 KB, 702 tokens by cl100k_base, as published. Nobody here has run it

Normalize Task Records (Inline Code)

When to use

Use this Skill when you need to normalize a list of task records and the executable logic is embedded directly in this document. Copy, adapt, and run the code block below rather than writing the logic from scratch.

Inputs

  • A JSON array of task records (in memory, or read from stdin).
  • Records may have inconsistent casing, whitespace, mixed status labels, and optional fields.

Expected output

  • A normalized list following the shared contract:
    • All string values trimmed.
    • id is a trimmed string.
    • status is trimmed, lowercased, then mapped (see code).
    • Missing optional fields are not added.
    • Key order per object: id, title, status, then remaining keys alphabetically.
    • Array sorted by id (numeric when all ids are integers, else lexicographic).

Procedure

Use the normalize function defined in the following code block. It is self-contained and deterministic — paste it into your runtime and call normalize(records).

def normalize(records):
    status_map = {
        "todo": "todo", "to do": "todo", "pending": "todo",
        "doing": "doing", "in progress": "doing", "wip": "doing",
        "done": "done", "complete": "done", "completed": "done",
    }
    key_priority = {"id": 0, "title": 1, "status": 2}

    def normalize_status(value):
        if not isinstance(value, str):
            return value
        key = value.strip().lower()
        return status_map.get(key, key)

    def ordered_keys(record):
        return sorted(record.keys(), key=lambda k: (key_priority.get(k, 3), k))

    def is_int(value):
        try:
            int(str(value).strip())
            return True
        except (TypeError, ValueError):
            return False

    if not isinstance(records, list):
        raise ValueError("input must be a JSON array")

    normalized = []
    for record in records:
        if not isinstance(record, dict):
            raise ValueError("each record must be a JSON object")
        out = {}
        for key in ordered_keys(record):
            value = record[key]
            if key == "status":
                out[key] = normalize_status(value)
            elif key == "id":
                out[key] = str(value).strip()
            elif isinstance(value, str):
                out[key] = value.strip()
            else:
                out[key] = value
        normalized.append(out)

    if normalized and all(is_int(r.get("id", "")) for r in normalized):
        normalized.sort(key=lambda r: int(str(r.get("id")).strip()))
    else:
        normalized.sort(key=lambda r: str(r.get("id", "")))
    return normalized

To run it end-to-end over stdin:

import json, sys
print(json.dumps(normalize(json.load(sys.stdin)), ensure_ascii=False, separators=(",", ":")))

Validation

  • normalize returns a list.
  • Every status is a mapped canonical value (or an intentionally unmapped label).
  • Records are ordered by id.
  • Non-list / non-object input raises ValueError.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 326,149. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.