Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add kmajerowicz/get-shit-right/plugin install gsrWrote 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/kmajerowicz/get-shit-right/verification)<a href="https://agentmods.dev/skills/kmajerowicz/get-shit-right/verification"><img src="https://agentmods.dev/badge/skills/kmajerowicz/get-shit-right/verification/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/kmajerowicz/get-shit-right/verification"><img src="https://agentmods.dev/badge/skills/kmajerowicz/get-shit-right/verification.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 | $0.00052 | $0.05191 |
| Opus 5.5 | $0.00021 | $0.02076 |
| Sonnet 5.5 | $0.00010 | $0.01038 |
| Haiku 4.5 | $0.00005 | $0.00519 |
Grade A, and why
gsr-verify 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 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.
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 — 457 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/gsr:verify — Verification Skill
You are executing the /gsr:verify command. Your job is to verify a completed feature or build phase against its must-haves and produce an evidence-based report.
Preview coverage and decisions
A preview or its visual approval is not proof of production behavior, all Truth pins or
human acceptance. Read current affected requirements and their single decision owner
(PRD/scope or focused feature Open Decisions). Compare preview assumptions/variant with
current wording/artifacts, including changed content under the same ID and missing files.
Use current evidence freshness and actual checks for integrated storage/transforms/etc.;
report simulation, unavailable browser observations and acceptance separately. Tests never
confirm D-pins. Later review-only/show-and-wait wins over recorded build scope. Follow
${CLAUDE_PLUGIN_ROOT}/docs/patterns/product-preview.md for the bounded transition.
Fresh-context stage handoff
At a major stage or long-build increment, follow
${CLAUDE_PLUGIN_ROOT}/docs/patterns/stage-handoff.md. Save objective/constraints,
original authorization/latest cadence, actual completed/remaining work, open choices,
current evidence/run limits, stage/checkpoint, exact next step/action and minimal read
references in the existing Active Task. Keep requirements in covered source documents.
Recommend /clear then dalej after saving. The next context loads coordinate and
the selected current guidance/refs only after the user's continuation; no automatic
hook execution, context reset or new session. Direct commands remain scoped optional
controls. A completed task or tiny correction needs no new-stage/reset instruction.
Planning-only completion records no authorized implementation next step. Partial/checked
build increments preserve actual overall in-progress state, limits and requested review.
Collaboration contract
Explicit task authorization and constraints override generic workflow defaults within host permissions. Read settled answers from the conversation and project files; never ask for them again. A planning-only request does not authorize implementation. Prepare plans, research material uncertainties, load installed skills, record state and run proportionate checks without separate approval. Weight controls rigor, not authority or review cadence. Honor explicitly requested per-chunk review and later instructions narrowing autonomy; otherwise continue reversible work in scope. Ask only for an unresolved consequential product choice or a real installation, spending, publishing, destructive-action or host permission boundary. Continue independent authorized work while a decision is pending. Completion reports give the outcome, brief verification and unresolved limits; detailed evidence stays accessible.
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 Changed · +31 lines e2254b417a55
- 2mo ago First seen · 426 lines · 52 tokens per session scan A b35b83b40abf
gsr-verify is a skill published in the GitHub repository kmajerowicz/get-shit-right (6 stars, last pushed 2d ago), licensed MIT. It adds 52 tokens to every session and 5,191 once invoked, about $0.0002 per session on Opus 5.5. 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.
Other skills, from other repositories
research-engineer
An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.
tika-eval-compare
Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".
cua-driver
Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.
jetson-validate-image
Use after jetson-flash-image to run static BSP checks, on-target smoke/regression tests on a flashed DUT, or both. Not for build or flash steps. Triggers: validate bsp, on-target validation.
atmos-validation
Validate Atmos projects, components, arbitrary JSON Schema inputs, EditorConfig, and GitHub Actions; use affected-file selection and native CI annotations.
skill-benchmark
Benchmark AI skill effectiveness by measuring implementation quality against legacy constraints.