I'm 15. I mass published 134K lines to hold AI agents accountable
Details
- External ID
- 47211441
- Source
- HN
- Company
- —
- Product
- I'm 15. I mass published 134K lines to hold AI agents accountable
- Website domain
- github.com
- Launched
- March 1, 2026
- Cohort
- —
- Upvotes
- 6
- Upvotes percentile
- 0.2853628536285363
- Tags
- —
- Fetched at
- Sept. 7, 2026, 9:26 p.m.
- Updated at
- Sept. 7, 2026, 9:26 p.m.
Enrichment
- Theme
- ai cybersecurity and penetration testing
- Vertical
- Horizontal
- Function
- Compliance & governance
- Audience
- B2B
- AI stance
- AI feature
- Project type
- Hobby / open-source project
- Normalized one-liner
- dataset for ai agent accountability
- Manually corrected
- False
Could you build this?
Yes Publishing large quantities of configuration, test specs, or audit wrapper scripts for AI agents is mostly prompt scripting and straightforward code generation.
Discussion
20 comments analyzed.
Competitors mentioned: LangChain, CrewAI, Vercel AI SDK, Smart contracts, CLI tools / MCP
Concerns raised: Cryptography doesn't verify actual code was running or enforcement logic was correct, No real non-repudiation without adversary in the system, Over-engineered solution in search of a problem, 134K lines of code mostly AI-generated with minimal human review, Unclear use case for verification if enforcement already happens pre-execution
Feature requests: Separate core modules (action-log, verifier, DSL, parser) from kitchen-sink tooling, Make compliance layer auditable and inspectable, Reduce to small trustable core instead of large codebase, Clarify multi-party use cases in examples, Lead with actual cryptographic kernel size (3-4K lines) not total LOC
Competitors
Other products that read as similar to this one — 62 launches clear the similarity bar, closest 8 shown.
Attention rank: #44 of 63 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 123 days after the earliest competitor.
- AI agent got 237 rules from another agent, still made the same mistakes · hn · 2026-03-24 · 6 upvotes · similarity 0.47
- I built an AI agent to yell at me about my ADHD · hn · 2026-06-30 · 6 upvotes · similarity 0.45
- I built a simulated AI containment terminal for my sci-fi novel · hn · 2026-02-17 · 23 upvotes · similarity 0.42
- I'm 15 and built a free tool for reading ancient texts. · hn · 2026-02-07 · 8 upvotes · similarity 0.41
- I tested 15 free AI models at building real software on a $25/year VPS · hn · 2026-04-02 · 37 upvotes · similarity 0.39
- AI scared me into building a career risk scorer (careerrisk.ee) · hn · 2026-03-30 · 6 upvotes · similarity 0.39
- I'm 15, built an AI that watches your screen and acts before you ask · hn · 2026-06-16 · 5 upvotes · similarity 0.39
- I submitted 316 AI-generated PRs to open source · hn · 2026-05-12 · 6 upvotes · similarity 0.38
Other launches for this product
- No other launches for this product.
Same idea, different domain
Nobody's really built a compliance & governance tool for Media & entertainment yet.