000-skill-loader-core

A loader for SKILL.md files, which are instruction files that teach an agent how to perform a task, with support for loading their summary, full text, or individual resources.

In plain words
What is it for?
Use it to discover available skills, read a skill’s metadata or complete instructions, and load selected reference files or templates.
Why use it?
It avoids loading every instruction and related file at once, so agents can retrieve only the level of detail they need.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/paulbreuler/limps/000-skill-loader-core
Clone the repo
git clone --depth 1 https://github.com/paulbreuler/limps
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 397 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.00397
Opus 5 $0.00000 $0.00198
Sonnet 5 $0.00000 $0.00079
Haiku 4.5 $0.00000 $0.00040

Measured 3d ago against content hash 9507fd90d218, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

000-skill-loader-core scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plans/0046-limps-skills/agents/000-skill-loader-core.md · 50 lines

What it actually says

Agent 000: Skill Loader Core

Objective

Build the core skill loading infrastructure that reads SKILL.md files from dist/skills/ with progressive disclosure support.

Tasks

  1. Define skill types (src/skills/types.ts)

    • SkillMetadata: name, description, version, tags, allowed-tools
    • SkillContent: full SKILL.md content
    • SkillResource: reference files, templates
  2. Implement SkillLoader class (src/skills/loader.ts)

    • listSkills(): Return metadata only (Tier 1)
    • loadSkill(name): Return full content (Tier 2)
    • loadResource(skill, path): Return specific resource (Tier 3)
    • Auto-discover skills from dist/skills/ directory
  3. Create initial skill (dist/skills/limps-planning/SKILL.md)

    • Follow Anthropic Agent Skills Standard format
    • Include frontmatter with required fields
    • Document core limps workflow

Acceptance Criteria

  • SkillLoader.listSkills() returns array of metadata objects
  • SkillLoader.loadSkill('limps-planning') returns full SKILL.md content
  • Skills directory structure follows Anthropic standard
  • Metadata extraction uses < 100 tokens per skill
  • Full skill content uses < 5k tokens

Technical Notes

  • Use gray-matter for frontmatter parsing
  • Skills are read-only (no write operations)
  • Path resolution must work in both dev and production
Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 50 lines · 0 tokens per session scan A 9507fd90d218

Subscribe to this mod's changes

000-skill-loader-core is an agent published in the GitHub repository paulbreuler/limps (10 stars, last pushed 6mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 397 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.