Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add crabsmadethis/d2r-horadric-tools --skill d2r-safe-editgit clone --depth 1 https://github.com/crabsmadethis/d2r-horadric-toolsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/crabsmadethis/d2r-horadric-tools/d2r-safe-edit)<a href="https://agentmods.dev/skills/crabsmadethis/d2r-horadric-tools/d2r-safe-edit"><img src="https://agentmods.dev/badge/skills/crabsmadethis/d2r-horadric-tools/d2r-safe-edit/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/crabsmadethis/d2r-horadric-tools/d2r-safe-edit"><img src="https://agentmods.dev/badge/skills/crabsmadethis/d2r-horadric-tools/d2r-safe-edit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00035 | $0.00799 |
| Opus 5 | $0.00017 | $0.00400 |
| Sonnet 5 | $0.00007 | $0.00160 |
| Haiku 4.5 | $0.00003 | $0.00080 |
Grade A, and why
d2r-safe-edit scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.
D2R Safe Edit
Overview
Every D2R save file edit follows the same cycle: backup, edit one thing, scan, verify, repeat. Skipping steps or batching edits has caused 15+ hours of debugging across multiple sessions. This skill is non-negotiable.
The Cycle
backup → edit ONE thing → scan → verify → next edit
Never batch. Never skip scanning. Never deploy with scanner errors.
Before Any Edit
-
Backup the .d2s (Rule 3)
shutil.copy2(path, path + '.pre_DESCRIPTION_bak')Verify the backup exists before proceeding.
-
Verify preconditions:
- UID exists in data file (Rule 1) —
d2r_lookup_uniqueor readunique_items.py - Item code matches base type in
item_bases.py - Target storage has free space (check grid with scanner)
- Stats are non-zero (Rule: Dex/Energy must never be 0)
- UID exists in data file (Rule 1) —
During Edit
-
Edit ONE thing. Not two. Not "one more small fix."
- Stats OR items OR skills — pick one per cycle
- If using chargen: one
--phaseat a time
-
Write to temp, verify, then overwrite (Rule 10)
- Never write directly to a live .d2s
- Parse the temp file back with scanner
- Verify item counts/positions/types match intent
- Only then copy to real location
After Edit
-
Run d2rdoctor scanner (Rule 4)
python3 -m d2r_chargen scan CHARNAME -
Evaluate ALL scanner output (Rule 19)
- Hard error → STOP. Fix before proceeding. Never dismiss as false positive without bit-level proof (Rule 17).
- Checksum mismatch → recalculate (Rule 5)
- LF inconsistent → fix merc state
- Encoding issues → rebuild item
-
Only proceed to next edit when scanner is clean.
Red Flags — STOP Immediately
- "Let me just do this other edit too" → NO. One edit per cycle.
- "That scanner warning is probably fine" → NO. Investigate every error (Rule 19).
- "I'll scan after all the edits" → NO. Scan after EACH edit (Rule 12).
- "The number makes semantic sense" → NOT proof. Check actual encoding (Rule 17).
- "Let me rebuild the whole file" → NO. Targeted section edits only (Rule 2).
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 87 lines · 35 tokens per session scan A eed1b541d913
d2r-safe-edit is a skill published in the GitHub repository crabsmadethis/d2r-horadric-tools (4 stars, last pushed 1mo ago), licensed MIT. It adds 35 tokens to every session and 799 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
assets-shader-get-data
Get detailed data about a shader asset — properties, subshaders, passes, compilation messages, and supported status. Supports token-saving path-scoped reads via paths or viewQuery. Use 'assets-find' with t:Shader or 'assets-shader-list-all' to locate the shader first.
gameobject-component-get
Get detailed information about a specific Component on a GameObject — type, enabled state, and (optionally) serialized fields and properties. Supports token-saving path-scoped reads via paths or viewQuery. Use 'gameobject-find' to list components first.
gameobject-component-modify
Modify a specific Component on a GameObject in opened Prefab or in a Scene. Allows direct modification of component fields and properties without wrapping in GameObject structure. Use 'gameobject-component-get' first to inspect the component structure before modifying. Three modification surfaces are available…
object-modify
Modify a Unity UnityEngine.Object's serializable fields/properties. Three modification surfaces are available (objectDiff, pathPatches, jsonPatch) — see the skill body. Use 'object-get-data' first to inspect the object structure.
reflection-method-call
Call a C# method by reflection — including private methods. Requires a method schema obtained via 'reflection-method-find'. Supports static methods, instance methods (with optional target deserialization), and main-thread / off-thread execution.
script-execute
Compiles and executes C# code dynamically using Roslyn. Supports a full-code mode (default) and a body-only mode — see the skill body for the difference and for how to pass Unity object references as parameters.