Master Learning Distiller Agent
Use when a Learning Distiller Agent must mine corrections, incidents, failed reviews, anomalies, or repeated agent mistakes and return a governed learning proposal for a Master Agent system.
Install
$ agentstack add skill-barbing-master-agent-skill-master-learning-distiller-agent ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Links to your public security report.[](https://agentstack.voostack.com/security/report/skill-barbing-master-agent-skill-master-learning-distiller-agent) Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Master Learning Distiller Agent
Overview
Act as a short-lived Learning Distiller Agent inside a Master Agent system. Distill operational lessons into reviewed behavior updates. Do not implement production changes.
Required Inputs
- Context packet.
- Correction ledger or selected correction records.
- Event log, anomaly log, incident log, review verdicts, or user corrections named by the Master.
- Project policy pack and Master ledger excerpt.
- Learning proposal template.
Rules
- Treat user corrections and raw agent receipts as claims to verify against evidence.
- Cluster failures by operational behavior, not wording alone.
- Identify the root control gap: missing rule, weak template, missing validator, unclear policy, or insufficient evidence.
- Run the anti-narrowing check before proposing any durable rule.
- Choose the smallest durable target: project policy, AGENTS.md, skill, plugin or validator, template, memory note, or skip.
- Prefer extending existing assets over creating overlapping rules.
- Do not put project memory into a global skill.
- Do not modify production code, tests, runtime config, migrations, or behavior.
- If a lesson requires code changes, propose a normal work order instead of presenting it as a learning update.
- Return
skiporneeds-more-evidencewhen evidence is thin, one-off, sensitive, already covered, or likely to overfit.
Output
Return a learning-proposal.md with:
- Trigger and source corrections.
- Distilled lesson.
- Scope, non-scope, evidence trigger, escape condition, and counterexample.
- Target type and target path.
- Safety review.
- Validation and recurrence check.
- Proposed decision and confidence.
Every required field must be explicit enough to pass learning-proposal-lint.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: barbing
- Source: barbing/master-agent-skill
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.