Owndiff
Verify human ownership of AI-assisted source-code changes by analyzing git diffs, generating easy diff-grounded multiple choice questions with the active agent model, and blocking risky pushes or merge requests until the human passes the ownership gate. Documentation and other non-source-only changes receive a report without multiple choice questions or gate artifacts.
Install
$ agentstack add skill-owndiff-own-your-diff-owndiff ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Links to your public security report.[](https://agentstack.voostack.com/security/report/skill-owndiff-own-your-diff-owndiff) Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
OwnDiff
Read and follow the complete OwnDiff instructions in ${CLAUDE_PLUGIN_ROOT}/SKILL.md.
Use the installed owndiff executable as the normal review flow. If owndiff is already available on PATH, reuse it. If it is missing, bootstrap the released CLI into ${OWNDIFF_BIN_DIR:-$HOME/.local/bin} with install.sh, set OWNDIFF_CMD to that executable path for the current workflow, and continue without asking the user to install the CLI manually first. Run "$OWNDIFF_CMD" run --repo . --out-dir .owndiff; each run starts a fresh review for the current diff and clears old local answers, gates, prompts, reports, and stale canonical LLM responses before writing current artifacts. When questions are pending, it starts a localhost server and opens the human's default browser. The browser review shows hints by default, lets the human retry before submitting, attempts to close itself after submission, then exits back to the same terminal session and can best-effort refocus known macOS terminal apps. Never launch the quiz in a detached/background terminal or second agent console. Never route the human to a separate multiple choice question command. Never substitute another checkout, use web search for question generation, print the answer key, or allow a push or merge request while the OwnDiff gate is blocked.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: owndiff
- Source: owndiff/own-your-diff
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.