agentsclimarketplace

Grill me

Skill s-frantz/agent-memory-trellis/skills/grill-me

The fundamental skill kit for agentic engineering with persisted memory.

Install
npx -y skills add s-frantz/agent-memory-trellis --skill grill-me

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

  • 1 stars1 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

Interviews you relentlessly about a plan or design until you reach shared understanding, resolving each branch of the decision tree. For each question it offers its recommended answer.

SKILL.md

1.9 KB, as published. Nobody here has run it

Interview me relentlessly about every aspect of this plan until we reach a shared understanding. Walk down each branch of the design tree, resolving dependencies between decisions one-by-one. For each question, provide your recommended answer.

Ask the questions one at a time.

If a question can be answered by exploring the codebase, explore the codebase instead.

Stop when the branches are resolved — when further questions would re-litigate settled decisions rather than uncover new ones. Say so explicitly.

Outputs

A grilling produces decisions, and decisions want to be written down — but this skill never silently edits canonical docs mid-interview. When the grilling concludes:

  1. Offer to write the resolved plan as an ephemeral working doc (e.g. docs/wip/PLAN-<topic>.md if the repo keeps a wip space, otherwise wherever drafts live), with date, purpose, and what would graduate it.
  2. Propose — don't apply — any updates the resolved plan implies for roadmap or decision docs. I, the user, must accept or decline each.

A grilling surfaces decisions; it does not enter them into the decision log. Where a resolved decision warrants a formal record, flag it in the PLAN as an ADR candidate (status: proposed) — nothing more. It graduates to the log (DECISIONS.md or a docs/decisions/ MADR file) only when the code reflects it — a promotion that happens later, during cleanup or a full audit, not here. A planned-but-unbuilt decision in the live log reads as in-force and misleads the next agent.

Be aware that invoking this skill can therefore result in new temporary files and doc updates. This is intended, but it happens with consent, at the end, never as a silent side effect.

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.