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 dannyhoang249-hub/pm-vietnamese-skill --skill pm-karpathy-thinkinggit clone --depth 1 https://github.com/dannyhoang249-hub/pm-vietnamese-skillWrote 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/dannyhoang249-hub/pm-vietnamese-skill/pm-karpathy-thinking)<a href="https://agentmods.dev/skills/dannyhoang249-hub/pm-vietnamese-skill/pm-karpathy-thinking"><img src="https://agentmods.dev/badge/skills/dannyhoang249-hub/pm-vietnamese-skill/pm-karpathy-thinking/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/dannyhoang249-hub/pm-vietnamese-skill/pm-karpathy-thinking"><img src="https://agentmods.dev/badge/skills/dannyhoang249-hub/pm-vietnamese-skill/pm-karpathy-thinking.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 | $0.00062 | $0.00753 |
| Opus 5 | $0.00031 | $0.00377 |
| Sonnet 5 | $0.00012 | $0.00151 |
| Haiku 4.5 | $0.00006 | $0.00075 |
Grade A, and why
pm-karpathy-thinking 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 — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Karpathy-inspired PM Thinking | Tư duy Karpathy cho PM
Apply four behavior principles—adapted from multica-ai/andrej-karpathy-skills—to product decisions. Do not copy software implementation guidance into a PM artifact; use the principles to improve decision quality.
Language | Ngôn ngữ
Match the requester’s language. For a bilingual audience or an explicit VI/EN request, use paired Vietnamese-English headings. Translate decisions, assumptions, success criteria, risks, and action items; preserve names, dates, evidence, metrics, URLs, event names, and identifiers exactly.
Four principles | Bốn nguyên tắc
1. Think before acting | Nghĩ trước khi làm
- State facts, assumptions, unknowns, and decision constraints separately.
- Surface material alternative interpretations rather than silently selecting one.
- Name the trade-off and push back if the requested solution does not address the stated problem.
- Ask only questions that would change the recommendation; otherwise proceed with labeled assumptions.
2. Simplicity first | Đơn giản trước
- Recommend the smallest viable scope, experiment, analysis, or release that can test the outcome now.
- Do not add speculative features, segments, metrics, processes, dashboards, or future-proofing.
- Prefer an existing workflow, prototype, concierge test, or instrumented release before a large build when it can answer the decision.
- Ask: would an experienced PM call this scope overcomplicated? If yes, simplify.
3. Surgical scope | Phạm vi có chủ đích
- Trace every proposed requirement, task, and metric to an explicit user outcome, risk, or constraint.
- Do not turn a narrow request into a strategy reset, full roadmap, or expansive PRD.
- Flag adjacent opportunities and pre-existing issues; do not absorb them into the current commitment unless requested.
- Remove only work made unnecessary by the proposed change; preserve existing commitments unless the decision explicitly replaces them.
What ships with it
1 file 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.
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 · 61 lines · 62 tokens per session scan A 27499519cebb
pm-karpathy-thinking is a skill published in the GitHub repository dannyhoang249-hub/pm-vietnamese-skill (1 stars, last pushed 24d ago), licensed MIT. It adds 62 tokens to every session and 753 once invoked, about $0.0003 per session on Opus 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
working-days-get-working-days
Calculate the number of working days between two dates, optionally accounting for country-specific holidays.
speckit.status
Progress Dashboard - Displays project progress status.
speckit.taskstoissues
Issue Tracker Syncer - Synchronize tasks.md to issue tracker.
speckit.backlog
Backlog Manager - Manage Ideas, Pending Requests and scan TODO/FIXME from codebase.
speckit.roadmap
Roadmap Strategist - Manage high-level roadmaps (Milestones) and transitions between Phases.
fde
Keeps the engagement record for client work. Use when they name a client or stakeholder. Use when they debrief a meeting or paste notes. Use when they ask what was agreed. Use when they run a POC, change the client's codebase, prove it on their staging, go live, or need evals before a model acts. Use when they prep a…