The skill and llms.txt told an agent to write --dest=<card>/bios bare.
A folder now goes in single quotes, attached to its option, and one that
holds a quote, a backtick or a control character, or starts with -, is
a question for the user.
ARD reserves text/markdown for agent skills, and Lighthouse warned on
the llms.txt entry. skills/setup-bios/SKILL.md holds the steps that set
up a device with the fewest files, and the manifest lists it as a skill.
llms.txt and an ARD manifest name the pages and the JSON API, and lead
an agent setting up a device to --for-games --required-only. Every page
registers four read-only WebMCP tools: the setup command, a hash lookup,
an emulator's files and the site search.
--for-games --required-only on a dArkOS card from a computer kept every
file of the list: --dest skipped the card's detection, so its rom
folders were never found. The games part beside the BIOS folder now
names them. The manifest marks a data tree every core runs without
(optional), and --required-only leaves it out: one GBA and one SNES game
took FBNeo's cheats and samples, 3,472 files no game opened.
validate.yml runs export_native with continue-on-error, and pr_checks
failed the whole run on it. The step is now shown and does not fail
the run, and a test keeps that list equal to the workflow's.
Code moved to sitecollection.py and the triage progress added errors
the baseline did not hold. A reused loop name, two lists without a
type, returns without a value, and a regex match read without a test
for None. Two baseline entries no longer occur and are removed.
og:title and twitter:title carry the search title ("mGBA BIOS
files"), an emulator page shares its maker's card, a download page is
dated by its release in the sitemap, and a Switch emulator's files are
called keys.
recipes/archives.json records each of the 2,222 archives database.json lists, .variants/ copies included: members in order with size, crc32 and sha1, the container sha1, md5 and size, exact when the TorrentZip build gives its bytes (1,790) and canonical otherwise (432), the profile and DAT recipes agreeing on its members, and the platforms or profiles pinning its bytes.
romset_recipes.py --capture writes it (pipeline step 1c), --build and --build-all rebuild from held members, loose first, then another archive, then its own copy, read back and checked before the move into place. Under zlib-ng nothing is claimed exact. A test fails when an archive has no recipe or a member is held nowhere.
The importer keeps the archive names the collection holds and every set whose member list equals an archive the collection keeps under another name (aga.zip is isa_aga), under the collection's name with the upstream one in set. All 113 MAME listxml releases (0.177 to 0.289) are imported, not one in five, and FBNeo is read again at its tip: 2,936 MAME recipes and 145 FBNeo ones, compacted as before.
fccpu20.zip, fcisio1.zip and interpod.zip now have recipes and carry slot ROMs beside their own set. Trimmed, each equals the archive under the newer MAME name, so they are kept as they are until the storage decision.
write_torrentzip reads and deflates one member at a time, gives back what it read (size, crc32, sha1) and takes a ZIP64 field only past 32 bits, so a 2 GiB disk image builds without being held whole. TorrentZip layouts are unchanged. zlib_ng_version() names the deflater that cannot reproduce TorrentZip bytes.
Recalbox hashes a ZIP by its members (zip_digest: composite, Bios.cpp:64-68), so its md5 names no container bytes: romset_recipes --missing tried to rebuild it and counted it unreproducible.
validate.yml runs mypy against its baseline and checks that shellcheck
is installed, and pr_checks ran neither. The guard test only knew the
python, vermin and npm commands, so it missed both. A step can now pipe
its output into a second command, as mypy into mypy-baseline filter.
check_platform_folder.py, the script the site publishes, still accepted
a cut md5 by prefix. It now compares whole digests as Batocera does
(batocera-systems:1088), like verify since cut_md5_owner.
Six places read an md5 shorter than 32 digits, each with its own bound.
digest_verdict accepted any prefix down to one digit, while Batocera
compares the whole string (batocera-systems:1088). A cut value now names
a dump only when one held md5 starts with it (cut_md5_owner, read by the
scraper, the resolver and --from-md5), and every verdict compares whole
digests.
With no required firmware among the bundled entries the line read
'20 (20 more, --verbose lists them)'. It now reads '20 (--verbose lists
them)', and reading-reports.md quotes the run as it prints.
gitignored_paths read a missing file as an empty registry: nothing to
restore, nothing unsatisfied, exit 0 with every release asset absent.
The run now stops with the path it could not read.
_pin_verified returned an existing <sha1>/<name> link without checking
its inode. A link whose cache name was gone had one link left, and the
prune of the same download removed it after it was returned, so
fetch_large_file handed out a path that no longer existed. The link is
now replaced when it holds another inode, and a revision directory a
prune removed before the link is made again.
Only verify.py and generate_pack.py moved to the repository root. From
any other folder slots, region_audit, provenance_report and bundled_only
said database.json was missing and to rebuild it, identity and
list_platforms named a missing folder, and game_rules ended on a
FileNotFoundError traceback. work_from_repo_root moves to common.py and
the eight tools call it: defaults are read from the root, a path the
user typed from where it was typed.
_collect_emulator_extras recorded a losing claim's path without the
declared root, while the winner's was rewritten with it. On RetroDECK
verify --region kept Rustation and Mednafen STV copies of withdrawn
bios/ files in undeclared_files as packed (7 under japan, 4 under
north-america, 3 under europe).
_pack_version took the text after the last underscore of the pack name.
AmberELEC's prerelease-20261019_1835 read as 1835 and the page would
announce a version gap that does not exist. The site now removes the
platform part the builder writes (platform_pack_prefix).
--emulator handy --required-only removed the empty ZIP, printed the
path it had just removed and exited 1, while verify reported 0/0 OK and
exited 0. The builder now says no file is needed and succeeds.
verify.py --emulator fbneo --strict > log 2>&1 put the ERROR verdict on
the first line of the log: redirected, stdout is block-buffered and
stderr is not, and run_cli flushed stdout only after main returned.
run_cli now sets stdout to line buffering, so every tool it runs keeps
the two streams in the order they were written.
Geargrafx drops a 512-byte header before its crc32 (src/media.cpp:754-773
at 7c447b0b), and its profile lists both sizes. slots compared the crc32
of the whole file, so EmuELEC's headered syscard3.pce read as contested.
Without --verbose each missing core file was followed by a bare '[size]'
line that told the reader nothing. It now reads 'validates size', the
verbose wording without the expected values.
Abdessamad Derraz