Add SKILL.md front door (check-family tune-up Phase 2) - #11
Closed
chaoz23 wants to merge 2 commits into
Closed
Conversation
'dmcheck run session.jsonl' with the packaged default charter (which names no GM) crashed with KeyError: 'rule' — core.check() returned an error pseudo-finding lacking the 'rule' key, and cli's counts summary assumed every finding has one. Same defect class as the 0.5.x 'init --dm' bug: worked in tests because tests always pass the fixture charter. Fix the class: check() now raises ValueError so a finding without 'rule' can never exist; cli routes it through the existing exit-2 error lane (JSON on stderr); watch validates GM up front; mcp already converts the raise to a clean error response. Regression tests cover cli and watch via subprocess (no-traceback asserted). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Agent-bootstrap skill file: table-vocabulary triggers, exit contract with the no-false-accusation framing (silence = nothing provable, never nothing wrong), real worked example from the messy-session fixture, MUST/MUST NOTs (no scores, no narrative judgment, name the GM), cross-family workflows, stale-knowledge deltas. README link. Stacked on fix/no-gm-crash: the changelog's 'clean exit 2, no traceback' claim is true as of that commit. Acceptance-tested: fresh-context agent produced the correct run --gm --dice-bot invocation and relayed findings with evidence and charter citation, silence as silence. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Owner
Author
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Template from srdcheck PR chaoz23/srdcheck#70, instantiated for dmcheck. 127 lines (cap 150).
Stacked on #10 (no-GM crash fix) — the SKILL.md changelog tells agents the no-GM case is a clean exit 2, which is true as of that fix. Merge #10 first (or merge this; it contains #10's commit).
Acceptance test passed: fresh-context agent, given only SKILL.md: correct
dmcheck run session.jsonl --gm "Greta" --dice-bot "RollBot"(inferred BOTH flags from the MUST section), relayed an R2 finding with charter citation + evidence excerpt, and reported a clean run as 'nothing provable against the charter — not a certificate the session was flawless.' The no-false-accusation contract survived transmission.Preflight clean.
🤖 Generated with Claude Code