agentsclimarketplace

Bounded skill optimization

Skill w3rdist-creator/agent-intelligence-economy/skills/bounded-skill-optimization

Obsidian + Hermes agent intelligence economy kit: clippings harvest, skill optimization, token efficiency, verification, loop scoring.

Install
npx -y skills add w3rdist-creator/agent-intelligence-economy --skill bounded-skill-optimization

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

One thing to look at

  • 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.

What its author says it does

Copied from the file, not written here

Improve agent skills and runbooks through evidence-backed, bounded, validated patches.

SKILL.md

0.7 KB, 105 tokens by cl100k_base, as published. Nobody here has run it

Bounded Skill Optimization

Trigger after difficult sessions, repeated failures, useful source harvests, cron reviews, or stale skill behavior.

Loop: observed run evidence -> recurring success/failure pattern -> localized candidate edit -> validation gate -> accepted patch or rejected-edit record -> later review.

Acceptance criteria:

  • grounded in observed evidence;
  • reusable across future runs;
  • smaller than the problem it prevents;
  • belongs in this artifact;
  • not a stale one-off fact;
  • does not loosen safety or privacy boundaries.

What ships with it

Read from the repository

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

Keep looking

Skills are one crate of 328,083. 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.