Getting it into your agent
There is no command for this one: it runs only inside a plugin, and the catalogue could not identify which plugin ships it. The source is linked below.
Wrote 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/tarunlnmiit/inbox-to-action/verify-package-install)<a href="https://agentmods.dev/skills/tarunlnmiit/inbox-to-action/verify-package-install"><img src="https://agentmods.dev/badge/skills/tarunlnmiit/inbox-to-action/verify-package-install/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/tarunlnmiit/inbox-to-action/verify-package-install"><img src="https://agentmods.dev/badge/skills/tarunlnmiit/inbox-to-action/verify-package-install.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 | $0.00089 | $0.01721 |
| Opus 5 | $0.00044 | $0.00860 |
| Sonnet 5 | $0.00018 | $0.00344 |
| Haiku 4.5 | $0.00009 | $0.00172 |
Grade C, and why
verify-package-install scanned grade C 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
rm -rf "/Users/tarungupta/Making It Big/Claude/inbox-to-action/dist" How it starts
The opening of the file, as written. The whole thing — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verify inbox-to-action in a Clean Environment
Ground truth for this repo (do not guess):
- Package name on PyPI:
inbox-to-action(same as the repo dir) - Console script:
inbox-to-action→inbox_to_action.main:cli - Subcommands:
run | auth | mcp.run --mockis fully keyless (uses bundledfixtures/sample_inbox.jsonand never touches real Gmail).run(real) andauthneed Gmail OAuth client secrets;runalso needs an LLMPROVIDERunless usinghost/claude. - Extras:
[mcp](FastMCP server),[anthropic](Anthropic SDK),[dev](pytest, pytest-cov, pytest-mock, respx, ruff, mypy, build, twine) - Build backend: hatchling. CI (
.github/workflows/ci.yml) records a GitHub release on av*tag but does NOT auto-publish by default — publishing is via manualtwine(project-scoped token in~/.pypirc) unless repo varPYPI_TRUSTED_PUBLISH=trueenables OIDC. So a plain tag push is not a release. - Wheel bundles
inbox_to_action/fixtures/**([tool.hatch.build.targets.wheel]) —--mockdepends on it, so verifyingrun --mockalso proves the data shipped. - Repo path has spaces — quote every path, never
cdinside compound commands.
Security invariant to re-verify on every release
inbox-to-action cannot send email — Gmail scopes are readonly + compose only,
no send scope or send call anywhere. This is enforced by tests/test_gmail.py. Any
clean-env verification of the dev flow (step 5) must run that test and see it pass; a
release that breaks it must not ship.
Known recurring failures (check these first)
No module named ruff— ruff lives only in the[dev]extra. A plainpip install inbox-to-actionwill NOT have ruff; that is expected. Only expect ruff afterpip install -e ".[dev]".InvalidConfiguration: Malformedfrom twine — a broken~/.pypirc. Run the preflight below before anytwine upload.- Testing against stale wheels — always rebuild
dist/(or install from PyPI with an explicit==<version>) so you verify the artifact you think you're verifying.
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 · 132 lines · 89 tokens per session scan C 145d3255c5d1
verify-package-install is a skill published in the GitHub repository tarunlnmiit/inbox-to-action (2 stars, last pushed 1mo ago), licensed MIT. It adds 89 tokens to every session and 1,721 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
agentsop-domain-eval-set
Build and govern a 50-200 example domain-specific held-out benchmark sampled from real traffic. Distinct from public benchmarks (MMLU/HumanEval/GSM8K via lm-evaluation-harness) which measure GENERAL capability. Only a held-out domain set predicts whether THIS system works on YOUR data. Collect real examples, label…
agentsop-test-fix-loop
Decision protocol for wiring a verify-then-fix loop around a code-editing LLM agent. The agent edits → runs lint/test → reads the output → fixes → re-runs, bounded by an iteration cap and an escalation rule. Activates whenever a coder agent has a verifiable success criterion (exit code, type-checker output, failing…
agentsop-regression-gate
Build a held-out eval set, run it on every prompt/model change, and block regressions in CI. An LM change is a code change — gate it with a test suite (eval set + metric + threshold). Cross-framework SOP not surfaced by any single base skill.
frontend-testing
WHEN testing any front-end UI with DOM Testing Library; NOT for end-to-end browser journeys or visual-regression checks; behavior-first queries, userEvent flows, async patterns.
tdd
WHEN working in TDD Red-Green-Refactor; NOT ad-hoc coding; write tests first, add minimal code to green, then assess refactoring.
swift-testing
WHEN writing, running, or diagnosing Swift Testing suites, including migrating XCTest tests to them and a crashing or non-reporting test target under xcodebuild; NOT for authoring XCTest or XCUITest tests; returns macro-driven test patterns, the XCTest boundary, and the correct way to read xcodebuild results.