agentleFS
Sign inStart free
IgorGanapolskyIgorGanapolsky/ThumbGate/.agents/skills/activesaddler-curriculum-honesty/SKILL.md

activesaddler-curriculum-honesty

Skill29 starsChanged 7 months agoMITView on GitHub

ActiveSaddler (arXiv:2610.00906) non-stationary bandit curriculum steal. Dynamically abstracts agent mistake patterns into failure-pattern arms, reorders pre-tool evaluation for fail-fast execution (<0.5ms), and detects regression drift without external RL/PyTorch dependencies.

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Anti-Clone Boundary: NEVER install PyTorch, Ray, Stable-Baselines, or heavy RL packages.

    line 13
  2. Distinctive:

    Drift Detection: When a single failure pattern exceeds 60% of recent violations, elevate its interdiction priority and notify the …

    line 20

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

ActiveSaddler Dynamic Curriculum Honesty (FORMAT Steal)

Stolen from ActiveSaddler (Automated Curriculum Learning for Agent Harness Optimization, arXiv:2610.00906).

Core Posture

  • Verdict: COMPARE_NOT_CLONE.
  • Anti-Clone Boundary: NEVER install PyTorch, Ray, Stable-Baselines, or heavy RL packages. Keep all bandit math CPU-local and sub-millisecond.
  • Fail-Fast Rule Reordering: Evaluate rules with highest discounted failure risk first to terminate invalid actions in <0.5ms before secondary rules run.

Key Invariants

  1. Failure-Pattern Arms: Track pulls, violations, and discounted statistics per mistake pattern.
  2. Discounted UCB Policy: Apply decay factor $\gamma = 0.92$ to discount historical observations while exploring uncertain arms.
  3. Drift Detection: When a single failure pattern exceeds 60% of recent violations, elevate its interdiction priority and notify the reliability gateway.

Verification Commands

node scripts/activesaddler-curriculum-doctor.js
node scripts/activesaddler-curriculum-doctor.js --json
node scripts/activesaddler-curriculum-doctor.js --map-only
node --test tests/activesaddler-curriculum-doctor.test.js

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free