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 jain777/jobclaw-skills --skill request-human-inputgit clone --depth 1 https://github.com/jain777/jobclaw-skillsWrote 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/jain777/jobclaw-skills/request-human-input)<a href="https://agentmods.dev/skills/jain777/jobclaw-skills/request-human-input"><img src="https://agentmods.dev/badge/skills/jain777/jobclaw-skills/request-human-input/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/jain777/jobclaw-skills/request-human-input"><img src="https://agentmods.dev/badge/skills/jain777/jobclaw-skills/request-human-input.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 | $0.01173 |
| Opus 5.5 | $0.00032 | $0.00469 |
| Sonnet 5.5 | $0.00016 | $0.00235 |
| Haiku 4.5 | $0.00008 | $0.00117 |
Grade A, and why
request-human-input 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 2mo 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 — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.
request-human-input
Pause the loop, ask one clear thing, resume on reply. Hard rules: ../_shared/RULES.md.
Two modes
Compose mode (default — no --reply)
Decide if the situation needs human input. If yes, produce a mobile-first prompt + ≤ 3 options + a timeout default. Always writes requests/<id>.json so the the optional JobClaw agent's HITL MCP can pick it up.
Parse mode (--reply "<text>")
Match the reply to an option (case-insensitive prefix + fuzzy). Return a structured decision; mark the request file resolved.
Inputs (compose)
situation— 1–3 sentences. The blocker.policy— freeform string. Escalation rules (e.g., "Escalate any irreversible action; auto-proceed on captcha if MFA backup is configured").context_attachments[](optional) — file paths (screenshot, log) the human can review.request_id(optional) — caller-provided; else generate a 12-char hex slug.
Inputs (parse)
--request_id— which request the reply is for.--reply— the user's verbatim reply string.
Method (compose)
-
Decide if a human is needed. Apply the
policytext tosituation. Output{needs_human: bool, reason}. Iffalse, exit cleanly — no question, no file. -
If needed, compose the prompt (mobile-first):
- Stem — ≤ 1 sentence, ≤ 120 chars. The question itself.
- Options — 1–3 quick-reply labels, each ≤ 24 chars. Verb-led.
- Default on timeout — what JobClaw does if the human doesn't reply:
skip | queue | retry | abort. Defaultabortfor irreversible actions. - Context snippet — 1 line (e.g. "Captcha at Workday submit for Mercury, screenshot attached"). No more.
-
Persist. Always write
requests/<request_id>.json:{ "request_id": "...", "created_at": "ISO-8601", "situation": "...", "policy": "...", "prompt": "string ≤ 120 chars", "options": [{ "label": "string ≤ 24 chars", "value": "string" }], "timeout_default": { "after_minutes": 30, "action": "skip|queue|retry|abort" }, "context_refs": ["path/or/url"], "status": "open" }
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.
- 2mo ago First seen · 99 lines · 80 tokens per session scan A 0e8328cf97f8
request-human-input is a skill published in the GitHub repository jain777/jobclaw-skills (7 stars, last pushed 5mo ago), licensed MIT. It adds 80 tokens to every session and 1,173 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.
request-human-input compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| request-human-inputjain777/jobclaw-skills | Skill | 7 | 5mo ago | MIT | A | 80 | Claude Code |
| imagegenopenai/codex | Skill | 135k | today | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 100k | today | Apache-2.0 | A | 19 | Claude Code |
| browser-automationopenclaw/openclaw | Skill | 394k | today | MIT | A | 36 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | 4d ago | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 316k | 7d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 95 | Claude Code |