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 andr-ca/agentharness --skill requirements-clarificationgit clone --depth 1 https://github.com/andr-ca/agentharnessWrote 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/andr-ca/agentharness/requirements-clarification)<a href="https://agentmods.dev/skills/andr-ca/agentharness/requirements-clarification"><img src="https://agentmods.dev/badge/skills/andr-ca/agentharness/requirements-clarification/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/andr-ca/agentharness/requirements-clarification"><img src="https://agentmods.dev/badge/skills/andr-ca/agentharness/requirements-clarification.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 | $0.00082 | $0.01289 |
| Opus 5.5 | $0.00033 | $0.00516 |
| Sonnet 5.5 | $0.00016 | $0.00258 |
| Haiku 4.5 | $0.00008 | $0.00129 |
Grade A, and why
requirements-clarification 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.
How it starts
The opening of the file, as written. The whole thing — 162 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Requirements Clarification
Gather requirements before writing code. A few minutes of structured discovery prevents hours of rework. Use this skill when a request is ambiguous, underspecified, or likely to have non-obvious edge cases.
When NOT to clarify: The request is a single, well-scoped bug fix with one clear resolution. Don't block on questions you can resolve by reading the existing code.
Verify the premise before accepting the requirement
Not every requirement arrives as a conversation. Many arrive pre-formed — a filed issue, an audit finding, a review comment, a recommendation document — asserting that something is broken or missing. That is a claim, and claims can be stale, partly wrong, or right for the wrong reason. Check it against the current code before designing anything.
A requirement that arrives with a proposed solution is two claims:
- The problem exists.
- The proposed change addresses it.
Verify them separately — they fail independently, and the second fails more often. A real problem paired with a wrong remedy is the most expensive case, because the problem being real lends the remedy false credibility.
| What you find | Right response |
|---|---|
| Claim true, proposal fits | Implement it |
| Claim true, proposal wrong or backwards | Solve it another way; say why |
| Claim partly true | Narrow scope to the true part |
| Claim false or already fixed | Don't implement; record the evidence |
Record the null result. "I checked, and it's already correct" is a finding worth writing down. Without it the next person re-investigates from scratch, and a rejected recommendation is indistinguishable from an ignored one.
Watch for the check that passes for the wrong reason. When you write a check to confirm a claim, confirm the check actually exercises what it names. A verification that silently matches nothing looks exactly like one that found nothing wrong — a green result you cannot distinguish from "never ran" is not evidence.
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 · 162 lines · 82 tokens per session scan A 6a5b154aa57b
requirements-clarification is a skill published in the GitHub repository andr-ca/agentharness (1 stars, last pushed 2d ago), licensed MIT. It adds 82 tokens to every session and 1,289 once invoked, about $0.0003 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
webgl-holographic-foil
A self-contained WebGL2 hero: thin-film interference over a crushed-foil surface whose palette shifts with the viewing angle; move the cursor to tilt the film.
cdf-anomalygen
Use when running the CDF AnomalyGen pipeline over a defect dataset — mask placement, fine-tuning, synthetic defect-image generation (SDG), evaluation, quality refinement, and pseudo-labeling — even when the user names only one stage.
deepstream-sop
Use this skill when building, deploying, evaluating, debugging, or measuring latency for the DeepStream SOP Inference Microservice — a GPU-accelerated FastAPI service that detects whether operators perform assembly-line steps in order via event boundary detection (GEBD) plus VLM classification. Trigger even if the…
hsb-ip-packetizer
Choose or explain HSB Sensor RX packetizer fields for HOLOLINKdef.svh. Do not use for full defs, validation, or runtime APB programming.
parabricks
Plan NVIDIA Parabricks pbrun commands, translate CPU options and check runtime readiness. Use for individual tools, including planning-only questions.
jetson-diagnostic
Read-only Jetson health snapshot for identity, memory, GPU, thermal, power, storage, services, and top processes.