0.24.3 — the budget stops being estimated, and two skills stop naming a file they do not ship - #32
Merged
Merged
Conversation
… a file they do not ship Hygiene from the 2026-09-13 family audit (HK-11). - the body budget is MEASURED. This gate divided by a calibrated 3.9 chars/token and the calibration drifted: agent-evals estimated ~4961 and measures 4374, agent-orchestrator ~4762 against 4309 — two files reported past the working limit with 300+ tokens of real headroom. With no tokenizer the check discloses rather than issuing a verdict from the wrong instrument. - agent-harness and agent-interop named `test/validate.py`, a repository file that does not ship inside a skill; both say "this pack's repository validator" now - agent-harness declares compatibility — it ships scripts/audit_agent.py - plugin.json's homepage matches package.json and the forge - the README names what loads on demand (27 files, counted per skill from the tree — the first draft of that table was wrong on three of four rows) and both invocation forms; its installer line said "both skills" over four Gate: npm test EXIT=0; OK: agent-stack structurally valid (15 checks, v0.24.3). Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The body budget check now measures or discloses; a runner without tiktoken would disclose on every skill and gate nothing. Same step the skill-audit job has carried since make-skill v0.28.0, in the job that runs the repo validator. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Hygiene from the 2026-09-13 family audit (HK-11).
The one that matters
The body budget was estimated, and the estimate was wrong in a direction that fails good files. This gate divided by a calibrated 3.9 chars/token. Measured 2026-09-14 against the authority's tokenizer:
agent-evalsagent-orchestratorBoth were being reported past the 4750 working limit with 300+ tokens of real headroom. make-skill v0.28.0 closed exactly this for the family's CI auditor; this repo kept the divisor. With no tokenizer installed the check now discloses that it could not measure instead of judging.
The rest
agent-harness/SKILL.md:122andagent-interop/SKILL.md:44namedtest/validate.py— a repository file that does not ship inside a skill, so it resolves at the repo root and nowhere an installed reader stands. Both say "this pack's repository validator".agent-harnessdeclarescompatibility: it shipsscripts/audit_agent.py.plugin.json's homepage matchespackage.jsonand the forge.lssays 12/7/6/2, and three of four rows were wrong. Its installer line also said "both skills" over a pack of four.Verification
npm testEXIT=0 —OK: agent-stack structurally valid (15 checks = 11 named + 4 per-skill, 4 skill(s), v0.24.3).audit_skill.py --house: 0 GAP, 19 PASS.🤖 Generated with Claude Code