agentleFS
Sign inStart free
IgorGanapolskyIgorGanapolsky/ThumbGate/.agents/skills/wisdom-context-artifact-not-clone/SKILL.md

wisdom-context-artifact-not-clone

Skill29 starsChanged 7 months agoMITView on GitHub

Steal Wisdom.ai / Soham Mazumdar episode FORMAT: context as a reusable agent artifact (goal, constraints, sources, freshness, verifier), plus an explicit wrong-fit gate. Do NOT clone ACE, Foundry, or Snowflake OSI. Slash: /wisdom-context-artifact-not-clone.

Create a free account to mount
Files
SKILL.md

Wisdom.ai FORMAT — artifact, not SKU

Goal

Produce fail-closed agent-context packs for whom: ThumbGate agents assembling lessons/RAG — so context is a reusable artifact (goal, constraints, sources, freshness, verifier), consumer is agent, and wrongFit includes wisdom_ai.

Constraints

NEVERALWAYS
Clone Adaptive Context Engine / Foundry / OSInpx thumbgate agent-context-artifact --pack=
consumer=human / analyst warehouse packsconsumer: "agent"
Tribal knowledge as "the context"Tangible JSON artifact
Omit freshness or verifierCheckable asOf + runnable verifier
Skip the wrong-fit questionwrongFit includes wisdom_ai
Dual-edit six-block auditor / ContextFSComplement, don't replace

Reference

Examples (show, don't tell)

Weak: "The semantic layer is in the warehouse; agents will figure it out."

Gold:

$ npx thumbgate agent-context-artifact --json --pack=tests/fixtures/agent-context-artifact-gold.json --now=2026-09-16T13:00:00Z
ok: true  consumer=agent  wrongFit includes wisdom_ai

Procedures

npx thumbgate agent-context-artifact --json --pack=<pack.json>
npx thumbgate agent-context-artifact --clone-ace --json
npm run test:agent-context-artifact
  1. Require consumer=agent.
  2. Require goal, constraints, sources, freshness, verifier.
  3. Fail stale asOf vs maxAgeHours.
  4. Require wrongFit includes wisdom_ai.
  5. Refuse ACE/Foundry/OSI clones.

Rubric

  • gold pack → ok=true
  • human fixture → ok=false
  • --clone-ace → wisdom_clone_refused
  • doctor: npm run test:agent-context-artifact PASS
  • evidence: command output in the same turn

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