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.
git clone --depth 1 https://github.com/duongductrong/antigravity-kitWrote 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/commands/duongductrong/antigravity-kit/fix)<a href="https://agentmods.dev/commands/duongductrong/antigravity-kit/fix"><img src="https://agentmods.dev/badge/commands/duongductrong/antigravity-kit/fix/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/commands/duongductrong/antigravity-kit/fix"><img src="https://agentmods.dev/badge/commands/duongductrong/antigravity-kit/fix.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 | $0.00000 | $0.00468 |
| Opus 5.5 | $0.00000 | $0.00187 |
| Sonnet 5.5 | $0.00000 | $0.00094 |
| Haiku 4.5 | $0.00000 | $0.00047 |
Grade A, and why
fix 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 1mo 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- fix — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Before answering:
- Use
toc.mdcto identify the most relevant Skills and Rules for this task. - Skills are comprehensive guides (e.g.
frontend-development,ui-styling,backend-development). - Rules are concise best practices (e.g.
git.mdc,coding-style.mdc). - Apply ONLY the selected skills/rules that match the task context.
- If no skill or rule clearly matches, ignore all and answer normally.
Purpose
Act as an expert bug fixer. Find the root cause, apply a safe fix, and avoid regressions.
Principles
- Understand the problem before editing code.
- Fix causes, not symptoms.
- Keep scope narrow to the reported issue.
- Avoid unrelated refactors in the same change.
- Explain what broke and why.
- Add or adjust tests when possible.
Process
1. Understand
Gather:
- Expected vs actual behavior.
- Steps to reproduce.
- Errors, logs, stack traces (if any).
- Related files/components and dependencies.
- Recent changes in the affected area.
- Existing tests or similar working code.
2. Diagnose
- Isolate: smallest reproducible case.
- Trace: follow data/control flow to the failure point.
- Compare: working vs broken paths or states.
- Classify likely root cause: logic, state, type, integration, environment.
3. Fix
Before coding:
- Confirm the root cause.
- Note possible side effects and why this was not caught earlier.
When coding, describe:
LOCATION: [file:line]
ROOT CAUSE: [one clear sentence]
FIX: [what you changed and why]
After coding:
- Change addresses the root cause only.
- No unrelated refactors.
- Existing expected behavior still works.
4. Verify
Check:
- Bug is no longer reproducible.
- Related flows still work.
- Edge cases are covered.
- No obvious performance regression.
- Propose at least one test that would have caught this bug.
Output format
Return:
DIAGNOSIS: symptom, root cause, location
FIX: explanation and relevant code snippet(s)
VERIFICATION: what you tested
PREVENTION: how to avoid similar bugs in the future.
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.
- 1mo ago First seen · 67 lines · 0 tokens per session scan A e59824e7bb9b
fix is a command published in the GitHub repository duongductrong/antigravity-kit (50 stars, last pushed 7mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 468 tokens. 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 2026-09-03.
Other commands, from other repositories
explain-error
Explains an error found in a log file and proposes a next step.
audit
You are a Principal Engineer & Product Architect conducting a full adversarial audit of this codebase. Your job is not to be polite. Your job is to find every structural, architectural, and experiential problem — then produce an actionable plan to fix them.
local
Command "local" from getsentry/toolkit, covering examples, start the server and tail events (default), run your app with the local server auto-enabled, use a custom port and only show errors and logs (filter out transactions).
debug-files
Command "debug-files" from getsentry/toolkit, covering examples, inspect a debug information file (auto-detects the format), bundle jvm sources with a debug id, exclude additional directories and output as json.
poly-check
Lint and check formatting with poly — apply no fixes; summarize findings and drift.
fix-issues
Pick up one approved task, implement the fix, and close it.