Ghidra Re
OrbitCurve/firmware-reverse-engineering
Expert-level Ghidra reverse engineering for firmware binaries with emphasis on stripped binary analysis, automated function discovery, cryptographic routine identification, authentication logic…
Run and record a hardware experiment on the 2C53T bench using a controlled five-step cycle.
Install Bench Experiment
$ npx skills add DavidClawson/OpenScope-2C53T --skill bench-experiment -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install DavidClawson/OpenScope-2C53T bench-experiment --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/DavidClawson/OpenScope-2C53T.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/bench-experiment .claude/skills/bench-experiment && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "bench-experiment" agent skill from https://github.com/DavidClawson/OpenScope-2C53T/tree/main/.claude/skills/bench-experiment into .claude/skills/bench-experiment/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "bench-experiment", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/DavidClawson/OpenScope-2C53T/tree/main/.claude/skills/bench-experimentType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add DavidClawson/OpenScope-2C53T --skill bench-experiment -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install DavidClawson/OpenScope-2C53T bench-experiment --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/DavidClawson/OpenScope-2C53T.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.claude/skills/bench-experiment .agents/skills/bench-experiment && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "bench-experiment" agent skill from https://github.com/DavidClawson/OpenScope-2C53T/tree/main/.claude/skills/bench-experiment into .agents/skills/bench-experiment/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "bench-experiment", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add DavidClawson/OpenScope-2C53T --skill bench-experiment -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install DavidClawson/OpenScope-2C53T bench-experiment --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/DavidClawson/OpenScope-2C53T.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.claude/skills/bench-experiment .cursor/skills/bench-experiment && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "bench-experiment" agent skill from https://github.com/DavidClawson/OpenScope-2C53T/tree/main/.claude/skills/bench-experiment into .cursor/skills/bench-experiment/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "bench-experiment", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/DavidClawson/OpenScope-2C53T.git --path .claude/skills/bench-experiment--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add DavidClawson/OpenScope-2C53T --skill bench-experiment -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install DavidClawson/OpenScope-2C53T bench-experiment --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/DavidClawson/OpenScope-2C53T.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.claude/skills/bench-experiment .gemini/skills/bench-experiment && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "bench-experiment" agent skill from https://github.com/DavidClawson/OpenScope-2C53T/tree/main/.claude/skills/bench-experiment into .gemini/skills/bench-experiment/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "bench-experiment", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install DavidClawson/OpenScope-2C53T bench-experimentInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add DavidClawson/OpenScope-2C53T --skill bench-experiment -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/DavidClawson/OpenScope-2C53T.git skills-src && mkdir -p .github/skills && cp -r skills-src/.claude/skills/bench-experiment .github/skills/bench-experiment && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "bench-experiment" agent skill from https://github.com/DavidClawson/OpenScope-2C53T/tree/main/.claude/skills/bench-experiment into .github/skills/bench-experiment/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "bench-experiment", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add DavidClawson/OpenScope-2C53T --skill bench-experiment -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install DavidClawson/OpenScope-2C53T bench-experiment --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/DavidClawson/OpenScope-2C53T.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.claude/skills/bench-experiment .opencode/skills/bench-experiment && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "bench-experiment" agent skill from https://github.com/DavidClawson/OpenScope-2C53T/tree/main/.claude/skills/bench-experiment into .opencode/skills/bench-experiment/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "bench-experiment", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
bench-experiment- GitHub stars
- 116
- Token cost
- ~1k tokens
- SKILL.md length
- 589 words
- Files
- 1
- Skills in repo
- 1
- Repo updated
- First seen
- Licence
- GPL-3.0
At a glance
Run and record a hardware experiment on the 2C53T bench using a controlled five-step cycle.
- Works in 2 steps: **A negative result is worthless until a… → State what the test cannot see. If you…
- A bench measurement is about to be taken
- SKILL.md covers Why the extra fields, The cycle, Recording conventions and Git workflow, plus 1 more section
- Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
What it does
Bench Experiment is an agent skill from DavidClawson/OpenScope-2C53T. Run and record a hardware experiment on the 2C53T bench using a controlled five-step cycle. Use whenever a bench measurement is about to be taken, a hypothesis tested, or a prior result questioned.
Its SKILL.md is about 1k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in Security, covering Reverse engineering and malware. It works with Ghidra. The repository describes itself as: Open-source reverse engineering and firmware development for the FNIRSI 2C53T oscilloscope/multimeter/signal generator. The licence is GPL-3.0.
When your agent uses it
- A bench measurement is about to be taken
- A hypothesis tested
- A prior result questioned
Example prompts
- “/bench-experiment”
Workflow steps
2 steps, taken from the first numbered list in SKILL.md.
- **A negative result is worthless until a positive control has been shown to
- State what the test cannot see. If you cannot name a blind spot, you have
What it can do on your machine
Read from SKILL.md and the folder at commit 2eae008. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
No scripts in the folder and no shell commands in SKILL.md.
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Bench Experiment loads about 1k tokens when it runs. Until then it costs about 54 tokens; SKILL.md has 589 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from DavidClawson/OpenScope-2C53T at commit 2eae008, republished under its GPL-3.0 licence (© DavidClawson). 589 words, ~1,035 tokens.
Download SKILL.mdSave it as
.claude/skills/bench-experiment/SKILL.md (or your agent's skills folder).- name
- bench-experiment
- description
- Run and record a hardware experiment on the 2C53T bench using a controlled five-step cycle. Use whenever a bench measurement is about to be taken, a hypothesis tested, or a prior result questioned.
Bench experiment cycle
Every measurement on this project goes through the same five steps, plus two fields that exist because of how this project has actually failed.
Why the extra fields
This project's wrong answers have almost never come from bad hypotheses. They came from instruments that could not have detected what they claimed to exclude, and which returned a stable, plausible number anyway. A stable wrong number is indistinguishable from a right one.
Documented cases: SSPI status reads at /2 (garbage, believed for weeks);
PB4/MISO floating (every status read taken on a line with no defined idle
level); Exp F comparing output LEVELS, which cannot distinguish a pin driven LOW
from one left FLOATING, so it "excluded" a class it could not see; a fixed-bin
DFT that reported 1.44 where a peak search found 22.8; fpga scope range <n> 0
silently addressing BOTH channels because the argument is 1-based; usart tx
queueing into a task that a coldtrace build never creates, so the frames were
never transmitted at all.
Two rules follow, and they are not optional:
- A negative result is worthless until a positive control has been shown to work through the same path, in the same session. "X produced no signal" means nothing until something has produced a signal through that exact instrument.
- State what the test cannot see. If you cannot name a blind spot, you have not understood the instrument.
The cycle
Write one file per experiment in docs/experiments/, named
YYYY-MM-DD-NN-short-slug.md, using docs/experiments/TEMPLATE.md.
- Problem — the open question this serves. One sentence, and it must be a question the project actually has, not a restatement of the procedure.
- Hypothesis — a prediction with a falsifier: "if H is true we see A; if false we see B." If both outcomes are consistent with the hypothesis, stop and rewrite it.
- Procedure — exact commands, pin states, build flags, signal settings.
Include the preconditions you verified by readback rather than assumed
(register values,
S1:0347,CTRL1=0x0000202C, IDCODE anchor0x0120681B). - Control — the known-good case, run in the same session through the same path. Record its result first. If the control fails, the experiment is VOID, not negative, and must be recorded as such.
- Results — raw numbers, not impressions. Screen observations are weak evidence and must be labelled as such.
- Blind spots — what this test could not have detected.
- Conclusion — what is now established, what is excluded, and explicitly what is not. Prefer "void" over "negative" when the control did not hold.
Show full SKILL.md (171 more words)Show less
Recording conventions
- Withdraw in place, never delete. A withdrawn result stays in the file with
a
WITHDRAWN <date>: reasonbanner. Deleted mistakes get re-made. - Anchor FPGA measurements: read IDCODE (
0x11) at/256and confirm0x0120681Bbefore trusting any config-port read. - Never read Gowin STATUS (
0x41) during capture — it desynchronises a running configured part and only a true power cycle recovers it. - A true power cycle is POWER → "Goodbye" → unplug USB → replug. The pinhole reset and a POWER-button shutdown with USB attached both leave the FPGA powered.
Git workflow
- One branch per bench session:
bench/YYYY-MM-DD. - One commit per completed cycle, message
exp(NN): <one-line conclusion>, carrying the experiment file plus any firmware change it required. - At session end, write/extend the narrative post in
docs/devlog/and merge tomain. The devlog is the story;docs/experiments/is the evidence.
Scale to the question
A five-minute register poke does not need a full file — but it does still need the control. If you are about to record a negative, or to exclude something, write the file.
© DavidClawson, GPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in .claude/skills/bench-experiment of DavidClawson/OpenScope-2C53T.
Open the folder on GitHubat commit 2eae008
Compare with similar skills
Bench Experiment next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Bench Experiment this skillDavidClawson/OpenScope-2C53T | 116 | — | ~1k | Automated safety check: Pass | GPL-3.0 | |
| Ghidra ReOrbitCurve/firmware-reverse-engineering | 214 | — | ~4.2k | Automated safety check: Pass | Apache-2.0 | |
| Firmware Security ReportsOrbitCurve/firmware-reverse-engineering | 214 | — | ~4.1k | Automated safety check: Pass | Apache-2.0 | |
| Go Rust Reversezhaoxuya520/reverse-skill | 40k | 2 repos | ~339 | Automated safety check: Pass | MIT | |
| Reverse Engineering Signaturesvillith/relink-logs | 157 | — | ~7.5k | Automated safety check: Pass | MIT | |
| Analyzing Packed Malware With Upx Unpackermukul975/Anthropic-Cybersecurity-Skills | 34k | — | ~3k | Automated safety check: Pass | Apache-2.0 |
Similar skills
Firmware Security Reports
OrbitCurve/firmware-reverse-engineering
Evidence-based security report generation for firmware assessments.
Go Rust Reverse
zhaoxuya520/reverse-skill
A skill your agent uses for reverse engineering stripped Go and Rust binaries including runtime recognition, pclntab/moduel data recovery, panic strings, and idiomatic decompilation recovery.
Reverse Engineering Signatures
villith/relink-logs
A skill your agent uses when a Granblue Fantasy Relink game patch breaks the GBFR Logs hook — signatures no longer match, "Could not find match for pattern" / "Could not find <offset" warnings…
Analyzing Packed Malware With Upx Unpacker
mukul975/Anthropic-Cybersecurity-Skills
Identifies and unpacks UPX-packed malware samples, including binaries with modified UPX magic bytes or headers that block automated decompression, to recover the original executable for static…
Reverse Engineering Malware With Ghidra
mukul975/Anthropic-Cybersecurity-Skills
Reverse engineers malware binaries using NSA's Ghidra disassembler and decompiler to study internal logic, cryptographic routines, C2 protocols, and evasion techniques at the assembly and pseudo-C…
Related
Works with
Categories
Questions about Bench Experiment
What does Bench Experiment do?
Run and record a hardware experiment on the 2C53T bench using a controlled five-step cycle. Bench Experiment is an agent skill from DavidClawson/OpenScope-2C53T. Run and record a hardware experiment on the 2C53T bench using a controlled five-step cycle.
When should I use Bench Experiment?
Bench Experiment fits situations like: A bench measurement is about to be taken; A hypothesis tested; A prior result questioned.
How do I install Bench Experiment in Claude Code?
Run `npx skills add DavidClawson/OpenScope-2C53T --skill bench-experiment -a claude-code`. Or copy the skill folder (.claude/skills/bench-experiment in DavidClawson/OpenScope-2C53T) into .claude/skills/bench-experiment in your project. Claude Code loads it when a task matches its description.
How do I install Bench Experiment in Codex?
Run `npx skills add DavidClawson/OpenScope-2C53T --skill bench-experiment -a codex`. Or copy the skill folder (.claude/skills/bench-experiment in DavidClawson/OpenScope-2C53T) into .agents/skills/bench-experiment in your project. Codex loads it when a task matches its description.
Can I use Bench Experiment in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add DavidClawson/OpenScope-2C53T --skill bench-experiment -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/bench-experiment, .gemini/skills/bench-experiment, .github/skills/bench-experiment and .opencode/skills/bench-experiment in your project.
What does Bench Experiment need to run?
SKILL.md names no scripts, command-line tools or credentials: Bench Experiment is instructions for the agent only.
Does Bench Experiment access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Bench Experiment safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Bench Experiment use?
Bench Experiment is published under the GPL-3.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Bench Experiment use?
About 1k tokens (SKILL.md is roughly 4.1k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Bench Experiment?
Skills that share tags, products or a category with Bench Experiment: Ghidra Re (OrbitCurve/firmware-reverse-engineering, 214 stars), Firmware Security Reports (OrbitCurve/firmware-reverse-engineering, 214 stars), Go Rust Reverse (zhaoxuya520/reverse-skill, 40k stars) and Reverse Engineering Signatures (villith/relink-logs, 157 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Bench Experiment?
DavidClawson (a GitHub user) maintains it in DavidClawson/OpenScope-2C53T, which has 116 GitHub stars. The repository was last updated on October 8, 2026.
Source: DavidClawson/OpenScope-2C53T on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.