The new D2R stash has named tabs (PERSONAL/SHARED/GEMS/MATERIALS/RUNES)
where gems are stacked (one slot per type/tier) instead of individual
inventory grid positions. The old slot-by-slot inspect_area approach
found 0 matches because gem icons don't align to the 38x38 grid.
Fix: _gems_tab_pull_3() uses template_finder.search_all() to find gem
positions anywhere in the left panel, then ctrl+clicks the 3 best hits.
Cube must now be in the character's right inventory (not personal stash).
Verified: 101 flawless->perfect transmutes in one run. Batch cap 100->500.
Also adds _switch_to_gems_tab() and _open_cube_from_inventory() helpers.
Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
transmute.py: convert_all_gems_to_perfect() chains all four gem tiers
(chipped->flawed->standard->flawless->perfect) by calling the existing
_run_gem_transmutes_lod for each tier in order. Result gems from each
pass land back in stash and feed the next pass.
testbed.py: gems_all scenario triggers convert_all_gems_to_perfect with
stash open; also adds missing return 0 to _gems().
hermes.py: fix UnicodeEncodeError on cp1252 terminals when the LLM
response contains non-ASCII characters (arrows, bullets, etc.).
Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
tools/hermes.py grabs the current D2R window (same grab() path the bot
uses), sends it to the local Qwen vision model at 192.168.1.98:8010,
and runs an interactive REPL. The system prompt gives Hermes full context
on coordinate systems, template matching, input layer, and bot patterns
so it can reason about what's on screen and suggest feature code.
<botty-env-python> tools/hermes.py [initial prompt]
<botty-env-python> tools/testbed.py hermes [initial prompt]
Also wired into testbed.py as the 'hermes' scenario.
Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
- scripts/stash_inventory.py: fix BaseItem["dimensions"] being a list
not a dict — use dims[0] for height so the claimed-slot dedup works
and multi-slot items (Large/Grand Charms, 2h weapons) are no longer
double-counted
- scripts/make_stash_csv.py: new — convert stash_inventory.json to a
deduplicated stash_list.csv (name, page, stats) for trade reference
- stash_list.csv: current stash export (71 unique items)
- config/params.ini: add stash_scan_interval (default 0/off)
- src/config.py: parse stash_scan_interval from [general]
- src/bot.py: after stash+transmute, trigger scan every N runs when
stash_scan_interval > 0
- src/inventory/personal.py: reset transfer_failures per stash tab so
each tab gets its own 2-failure budget before moving to the next
- tools/testbed.py: PASS/FAIL fix for `stash all` — equipped-area cols
(≥4) correctly left in inventory is a PASS not a FAIL (Bug 19 guard)
- CLAUDE.md: document stash scanner scripts in File Quick Reference
Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
stash_all_items() treated any keep item remaining after transfer_items()
as "this tab is full", paged through every stash tab, and on the last page
called stash_full() -> taskkill D2R + a false Discord "stash full" alert.
A transfer can fail for non-fullness reasons (equipped-area click guard,
transient UI), so this could kill the game on a stash that has free slots.
Fix: before declaring a tab full, check EmptyStashSlot. If a slot is free
but the transfer still failed, count it as a transfer failure (cap 2) and
bail, leaving items in inventory -- never advance tabs / call stash_full()
on a non-full page. Also fix a leftover >3 page bound (-> >5) in the same loop.
Add a `stash` scenario to tools/testbed.py (self-bootstraps: main menu ->
create game -> walk -> open stash -> run stashing; `all` widens the scan to
all 10 columns to exercise the keep-item branch) and extract the conda-env
SSL cert workaround into a shared _patch_ssl() helper. Document as Bug 19.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
