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 shivamsriva31093/gcp-ironclad --skill gcp-dispute-kitgit clone --depth 1 https://github.com/shivamsriva31093/gcp-ironcladWrote 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/shivamsriva31093/gcp-ironclad/gcp-dispute-kit)<a href="https://agentmods.dev/skills/shivamsriva31093/gcp-ironclad/gcp-dispute-kit"><img src="https://agentmods.dev/badge/skills/shivamsriva31093/gcp-ironclad/gcp-dispute-kit/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/shivamsriva31093/gcp-ironclad/gcp-dispute-kit"><img src="https://agentmods.dev/badge/skills/shivamsriva31093/gcp-ironclad/gcp-dispute-kit.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 | $0.00080 | $0.02784 |
| Opus 5 | $0.00040 | $0.01392 |
| Sonnet 5 | $0.00016 | $0.00557 |
| Haiku 4.5 | $0.00008 | $0.00278 |
Grade A, and why
gcp-dispute-kit scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
RESP=$(curl -s -G "https://monitoring.googleapis.com/v3/projects/${P}/timeSeries" \ How it starts
The opening of the file, as written. The whole thing — 218 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GCP Dispute Kit (READ-ONLY)
Overview
Assembles everything a fraud victim needs to file: an exhibit-backed evidence
summary, ready-to-file letters for their situation and jurisdiction (Google +
India + US tracks), and a filing-sequence README with a deadline tracker.
Formalizes the practice in docs/incident-story.md. Mutates no cloud state.
Three entry states, three different products:
| State | Meaning | You get |
|---|---|---|
BLEEDING |
Spend is spiking right now | checklists/emergency-stop.md — nothing else until the bleeding stops |
FRESH |
Incident over, nothing filed | Full packet + Google console-dispute letter (+ India bank-chargeback letter when jurisdiction includes IN — the card-dispute clock runs from the statement date, not from Google's reply) |
STUCK |
Dispute filed but stalled | Packet + escalation letters (Google reply + IN/US external tracks) |
When to Use
- After a leaked-key incident, once spend is back to normal, to prepare the dispute.
- When an existing dispute is stalled: ghosted, partial offer, denied, or a refund that never landed.
- NOT while spend is actively spiking — the kit will hand you the emergency-stop checklist and halt.
Inputs
SESSION_DIRenv var (optional): defaults to/tmp/gcp-dispute-kit/$(date -u +%Y-%m-%dT%H-%M-%SZ)/.- Intake answers from the victim (Phase 0). Everything else is discovered.
Outputs
${SESSION_DIR}/packet/ — README.md, evidence-summary.md, letters/, exhibits/, manifest.json (validated against this skill's output.schema.json).
Safety
- Every command in this skill is read-only (
list/describe/query). Noupdate,enable,disable,create, ordeleteverbs anywhere except insidechecklists/emergency-stop.md, which the victim runs by hand in a live emergency. - A letter with an unresolved
{{placeholder}}is never emitted —scripts/render.pyhard-errors instead. - Claims are confidence-marked: billing-export-backed > CSV-backed > victim-stated. Never present a victim-stated number as data-backed.
What ships with it
27 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.
- checklists/emergency-stop.md 1.2 KB
- output.schema.json 2.7 KB
- scripts/analyze_billing.py 3.7 KB runs code
- scripts/classify_state.py 2.2 KB runs code
- scripts/render.py 1.8 KB runs code
- templates/evidence-summary.template.md 1.6 KB
- templates/google/console-dispute.template.md 2.6 KB
- templates/google/escalation-reply.template.md 2.1 KB
- templates/india/bank-chargeback.template.md 2.0 KB
- templates/india/cert-in-report.template.md 1.7 KB
- templates/india/nch-complaint.template.md 2.0 KB
- templates/packet-README.template.md 1.8 KB
- templates/placeholders.json 3.2 KB
- templates/us/ftc-complaint.template.md 2.0 KB
- tests/conftest.py 627 B runs code
- tests/fixtures/intake_fresh.json 73 B
- tests/fixtures/intake_stuck.json 78 B
- tests/fixtures/make_fixture.py 1.9 KB runs code
- tests/fixtures/manifest_sample.json 1.1 KB
- tests/fixtures/synthetic_billing.csv 13 KB
- tests/fixtures/values_fresh.json 1.8 KB
- tests/test_analyze_billing.py 2.6 KB runs code
- tests/test_classify_state.py 2.3 KB runs code
- tests/test_end_to_end.py 3.5 KB runs code
- tests/test_manifest_schema.py 772 B runs code
- tests/test_render.py 2.2 KB runs code
- tests/test_templates_lint.py 3.0 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.
- 12d ago First seen · 218 lines · 80 tokens per session scan A 9ef61ea27788
gcp-dispute-kit is a skill published in the GitHub repository shivamsriva31093/gcp-ironclad (24 stars, last pushed 1mo ago), licensed MIT. It adds 80 tokens to every session and 2,784 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
bionic-safety-net
Unified survival infrastructure: health, finance, legal safety, circuit breakers, and structural protection against all ruin classes. The last line of defense.
insurance-expert
Expert-level insurance systems, underwriting, claims processing, actuarial analysis, risk assessment, and insurtech solutions. Use when the user mentions underwriting, claims, actuarial, risk, or insurtech, or when the task involves Insurance Systems, Insurance Types, Standards and Regulations, or Claims Processing.
finops-expert
Expert-level cloud financial operations, cost optimization, and cloud economics. Use when the user mentions cloud cost, optimization, cloud economics, or AWS cost, or when the task involves FinOps Fundamentals, Cost Management, FinOps Practices, or Cost Visibility.
banking-expert
Expert-level banking systems, core banking, regulations, and banking technology. Use when the user mentions finance, core banking, regulations, or payments, or when the task involves Banking Systems or Key Technologies.
carta-witness-signatures
Status of witness and spousal-consent signature requests on option grants, RSAs, PIUs — who still needs to sign, and what's awaiting signature, signed, or expired. Covers one award, one company, or a whole portfolio. Read-only.
carta-hmrc-limits
Report a UK company's HMRC EMI/CSOP limit usage. Use when asked how much EMI allowance is left, whether the company is near the £6m EMI limit, who is close to or over their £250k individual limit, CSOP headroom, or whether a planned grant fits.