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 Reedtrullz/codex-antigravity-auth --skill antigit clone --depth 1 https://github.com/Reedtrullz/codex-antigravity-authWrote 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/reedtrullz/codex-antigravity-auth/anti)<a href="https://agentmods.dev/skills/reedtrullz/codex-antigravity-auth/anti"><img src="https://agentmods.dev/badge/skills/reedtrullz/codex-antigravity-auth/anti/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/reedtrullz/codex-antigravity-auth/anti"><img src="https://agentmods.dev/badge/skills/reedtrullz/codex-antigravity-auth/anti.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 | $0.11831 |
| Opus 5.5 | $0.00000 | $0.04732 |
| Sonnet 5.5 | $0.00000 | $0.02366 |
| Haiku 4.5 | $0.00000 | $0.01183 |
Grade A, and why
anti 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 23d 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 — 441 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Anti
Experimental PCM WAV listening input requires explicit local files,
an explicit Gemini model and --probe-unverified-audio before upload. Never treat
a text response, filename or transcript as proof of listening; actual backend
audio acceptance remains unverified. No live probe is implied by this skill.
Explicit local PNG/JPEG attachments use repeatable --image PATH and require
gateway-declared image support at every stage. Read attachments
for limits, unscanned pixels/metadata, retention, and bounded real-media
evaluation evidence. Do not infer image support from a model name or replace images with text.
Use offline benchmark replay for controlled synthetic quality evaluation. Ordinary compare output, model agreement and latency do not establish quality. Benchmark replay makes no model calls and never changes routing defaults.
Use this skill to ask the local codex-antigravity-auth gateway for an external Antigravity review, consult, deep work plan, named workflow preset, or bounded multi-model panel while native Codex remains the primary agent.
V3's primary product is native Claude in Codex through codex-antigravity setup; $anti is an optional helper for review and planning after the gateway and Codex model picker are already working.
Core Rule
Treat Antigravity output as a second opinion. Run the helper, read the result, then synthesize it with your own analysis before answering the user. Do not blindly forward the Antigravity result as final truth.
Literal @anti is a text convention in v1, not a guaranteed app-level mention chip. $anti is the reliable explicit skill invocation.
Panel, MoA, and Fusion workflows are advisory only. The helper can fan out to multiple gateway-advertised models and ask a judge model to synthesize their views, but Codex remains the acting agent and must verify findings before editing. Structured panel findings include a verify hint; run or inspect that local check before acting on the claim.
What ships with it
60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- agents/openai.yaml 357 B
- ARTIFACTS.md 9.7 KB
- ATTACHMENTS.md 5.7 KB
- BENCHMARK.md 7.5 KB
- CHUNK_RESUME.md 5.6 KB
- CONTEXT_PREFLIGHT.md 5.8 KB
- DATA_POLICY.md 5.8 KB
- DIFF_PROVENANCE.md 3.7 KB
- LOCAL_ONLY.md 5.6 KB
- MUSIC_REVIEW.md 9.3 KB
- OWNERSHIP.md 3.3 KB
- RUN_CONTROL.md 4.5 KB
- schemas/benchmark-adjudications-v1.json 3.0 KB
- schemas/benchmark-replay-v1.json 9.2 KB
- schemas/benchmark-report-v1.json 2.3 KB
- schemas/music-evidence-v1.json 3.0 KB
- schemas/music-review-compact-v1.json 2.1 KB
- schemas/music-review-v1.json 1.8 KB
- schemas/raw-lane-v1.json 512 B
- schemas/review-report-v1.json 5.2 KB
- schemas/run-index-v1.json 9.7 KB
- schemas/saved-result-v2.json 9.0 KB
- SCOPES.md 3.9 KB
- scripts/anti_lib/__init__.py 54 B runs code
- scripts/anti_lib/artifacts.py 18 KB runs code
- scripts/anti_lib/benchmark.py 20 KB runs code
- scripts/anti_lib/capabilities.json 30 KB
- scripts/anti_lib/capabilities.py 4.3 KB runs code
- scripts/anti_lib/checkpoints.py 16 KB runs code
- scripts/anti_lib/chunking.py 4.9 KB runs code
- scripts/anti_lib/cleanup.py 10 KB runs code
- scripts/anti_lib/console.py 1.1 KB runs code
- scripts/anti_lib/context_budget.py 7.0 KB runs code
- scripts/anti_lib/context.py 16 KB runs code
- scripts/anti_lib/data_policy.py 11 KB runs code
- scripts/anti_lib/diff_snapshot.py 12 KB runs code
- scripts/anti_lib/endpoint_policy.py 5.2 KB runs code
- scripts/anti_lib/errors.py 103 B runs code
- scripts/anti_lib/file_protection.py 6.8 KB runs code
- scripts/anti_lib/finding_verdicts.py 5.6 KB runs code
- scripts/anti_lib/inventory.py 9.2 KB runs code
- scripts/anti_lib/ledger.py 831 B runs code
- scripts/anti_lib/local_policy.py 5.7 KB runs code
- scripts/anti_lib/media.py 8.3 KB runs code
- scripts/anti_lib/music_evidence.py 10 KB runs code
- scripts/anti_lib/namespaces.py 2.6 KB runs code
- scripts/anti_lib/persistence.py 3.4 KB runs code
- scripts/anti_lib/redaction.py 5.0 KB runs code
- scripts/anti_lib/reflections.py 13 KB runs code
- scripts/anti_lib/reports.py 11 KB runs code
- scripts/anti_lib/retention.py 9.0 KB runs code
- scripts/anti_lib/review_corpus.json 4.9 KB
- scripts/anti_lib/run_control.py 4.2 KB runs code
- scripts/anti_lib/run_records.py 22 KB runs code
- scripts/anti_lib/runner.py 470 B runs code
- scripts/anti_lib/secret_redaction.py 12 KB runs code
- scripts/anti_lib/spend_control.py 13 KB runs code
- scripts/anti_lib/verifier.py 17 KB runs code
- scripts/anti_lib/wav_audio.py 21 KB runs code
- scripts/anti_lib/windows_file_security.py 13 KB runs code
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.
- 23d ago Changed · +1 lines 65fa53bd7c11
- 26d ago Changed · +7 lines 1d25e1e80bb4
- 28d ago Changed · +15 lines a78cbcca48e1
- 1mo ago Changed · +51 lines dc26d3d4e9b3
- 1mo ago Changed · +19 lines d57432ccc1e2
- 1mo ago Changed · +7 lines 68cd36265052
- 1mo ago Changed · +3 lines 092d3c8976d4
- 1mo ago Changed · +7 lines eca733d54230
- 2mo ago Changed · +3 lines c5171adb36ff
- 2mo ago First seen · 328 lines · 0 tokens per session scan A 7c51c40143ff
anti is a skill published in the GitHub repository Reedtrullz/codex-antigravity-auth (2 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 11,831 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 2026-08-31.
anti compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| antiReedtrullz/codex-antigravity-auth | Skill | 2 | yesterday | MIT | A | 0 | Codex |
| sovereign-economics-enginewinstonkoh87/Athena-Public | Skill | 620 | yesterday | MIT | A | 43 | Claude Code, Codex |
| social-physics-filterwinstonkoh87/Athena-Public | Skill | 620 | yesterday | MIT | A | 34 | Claude Code, Codex |
| therapeutic-ifswinstonkoh87/Athena-Public | Skill | 620 | yesterday | MIT | A | 34 | Claude Code |
| agentic-code-orchestratorwinstonkoh87/Athena-Public | Skill | 620 | yesterday | MIT | A | 42 | Claude Code, Codex |
| dashboard-builderwinstonkoh87/Athena-Public | Skill | 620 | yesterday | MIT | A | 31 | Claude Code |
| bionic-decision-enginewinstonkoh87/Athena-Public | Skill | 620 | yesterday | MIT | A | 41 | Claude Code, Codex |