HNHacker News
TopNewBestAskShowJobs

CuriouslyC

8,233 karma · joined July 30, 2014

Chief scientist/ceo, Sibylline Software. nathan @ sibylline.dev
submissionscomments

I Pivoted to Game Development

sibylline.dev·4 pts·CuriouslyC·
0

I Changed My Mind About MCP

sibylline.dev·3 pts·CuriouslyC·
0

You Don't Need to Detect Prompt Injection to Stop It

sibylline.dev·2 pts·CuriouslyC·
0

Smith: The Secure Open Source Multi-User AI Assistant Framework

github.com·4 pts·CuriouslyC·
2

Defeating Prompt Injection with Protocol Firewalls

sibylline.dev·2 pts·CuriouslyC·
0

The Pillars of Agent Security

sibylline.dev·2 pts·CuriouslyC·
0

The Pillars of Agentic Security

sibylline.dev·1 pts·CuriouslyC·
0

Clean: High Performance Prompt Injection Detection and Mitigation

github.com·1 pts·CuriouslyC·
1

GLM-5: Targeting complex systems engineering and long-horizon agentic tasks

z.ai·484 pts·CuriouslyC·
520

Scurl: Agent First Curl Wrapper with Markdown Extraction and Secret Blocking

github.com·1 pts·CuriouslyC·
1

Pi Is the Linux of Agent Harnesses

sibylline.dev·2 pts·CuriouslyC·
0

Averting the Code Quality Apocalypse

sibylline.dev·3 pts·CuriouslyC·
0

Stop screwing around with agent orchestration, your bottleneck is validation

sibylline.dev·2 pts·CuriouslyC·
1

The Problems with Spec Driven Development

sibylline.dev·1 pts·CuriouslyC·
0

Agents with Scribe Solve Hard Problems 57% More Often

sibylline.dev·1 pts·CuriouslyC·
0

Stop screwing around with agent orchestration, your bottleneck is validation

sibylline.dev·2 pts·CuriouslyC·
1

Scribe Increases Agent Success Rate on Hard Problems by 57%

sibylline.dev·1 pts·CuriouslyC·
0

Scribe reduces agent token usage by 30% with no loss of accuracy

sibylline.dev·2 pts·CuriouslyC·
0

Scribe reduces SWE-bench token usage by 30% with no loss of accuracy

sibylline.dev·1 pts·CuriouslyC·
0

Is your codebase holding back your AI tools?

codehealth.sibylline.dev·1 pts·CuriouslyC·
1

Digital Alchemy: Turning Slop into Gold with Ralph and Valknut

sibylline.dev·1 pts·CuriouslyC·
0

The Code Quality Leaderboard

codehealth.sibylline.dev·1 pts·CuriouslyC·
0

Model Evals Can Fix Education

sibylline.dev·2 pts·CuriouslyC·
0

Valknut: Code Intelligence for Agents

sibylline.dev·2 pts·CuriouslyC·
0

Valknut: Advanced Code Intelligence

sibylline.dev·1 pts·CuriouslyC·
0

Show HN: Valknut – static analysis to tame agent tech debt

github.com·2 pts·CuriouslyC·
0

Calligraphers and Storytellers

sibylline.dev·2 pts·CuriouslyC·
0

OpenAI Should Make a Phone

ft.com·3 pts·CuriouslyC·
2

Claude Skills Considered Harmful

sibylline.dev·4 pts·CuriouslyC·
0

AI Is Too Big to Fail

sibylline.dev·8 pts·CuriouslyC·
3
Page 1 of 2Next →