Fable holds only for Willison, Opus 4.8 only in Claude Code, map only as ASCII
Every URL the pipeline pulled into ranking for this issue — primary sources plus the supporting and contradicting findings each Researcher returned. Inline citations in the issue point back here.
Sources
sqlite-utils 4.0rc2, mostly written by Claude Fable (for about $149.25) simonwillison.net
I wrote about the sqlite-utils 4.0rc1 release a couple of weeks ago. Since we only have Claude Fable on our Max subscriptions for a few more days, I decided to see if it could help me get to a 4.0 stable release that I felt truly comfortable about, since I try to keep to SemVer and like my incompatible major versions to be as rare as possible. I started with this prompt, in Claude Code for web on my iPhone: Final review before shipping a stable 4.0 release - very important to spot any last minu…
sqlite-utils 4.0rc2 simonwillison.net
Release: sqlite-utils 4.0rc2 See sqlite-utils 4.0rc2, mostly written by Claude Fable (for about $149.25) .
Better Models: Worse Tools simonwillison.net
Better Models: Worse Tools Armin reports on a weird problem he ran into while hacking on Pi: The short version is that newer Claude models sometimes call Pi’s edit tool with extra, invented fields in the nested edits[] array. And not Haiku or some small model: Opus 4.8. The edit itself is usually correct but the arguments do not match the schema as the model invents made-up keys and Pi thus rejects the tool call and asks to try again. That alone is not too surprising as models emit malformed to…
Building a World Map with only 500 bytes simonwillison.net
Building a World Map with only 500 bytes Iwo Kadziela (assisted by Codex) figured out a way to generate a credible ASCII world map using 445 bytes of data: The key trick is to use deflate compression, which is then wired together using this neat snippet of JavaScript. I didn’t know you could use fetch() with data: URIs like this: fetch(‘data:;base64,1ZpLsgIxCEXnrM…==‘).then( r => r.body.pipeThrough(new DecompressionStream(‘deflate-raw’)) ).then( s => new Response(s).text() ).then( t => b.inne…
References
CloudZero — Claude Code agents/cost tracking cloudzero.com
Pro and Max subscribers are billed a flat monthly fee but subject to hidden 5-hour rate limits… AgentsView allows these subscribers to apply hypothetical API pricing to their local logs, providing visibility into their ‘burn rate’.
Futurum Group — independent review layer for coding agents futurumgroup.com
Having an AI model review its own output is more akin to a ‘spell-checker reviewing its own autocorrect’ than a rigorous audit… over 55% of decision-makers identify agent hallucination and reliability as their primary obstacle to full automation.
arXiv study of 567 agent-assisted PRs arxiv.org
83% were eventually merged, [but] nearly 45% required manual human revision to correct logic errors or satisfy project standards.
SoftwareSeni — AI-generated contributions and OSS supply chain softwareseni.com
The Jazzband collective, a prominent Python project hub, was forced to shut down in 2026, citing an unsustainable flood of AI-generated spam PRs… curl terminated its six-year bug bounty program in early 2026 after being overwhelmed by a wave of plausible-sounding but fundamentally invalid AI security reports.
Medium — Petra Ivanigova, ‘Why AI will make software more expensive’ medium.com
Code churn has doubled and copy-pasted code has risen by 48% in AI-assisted environments… 45% of AI-generated code samples introduced OWASP Top 10 vulnerabilities.
Substack — ‘I replaced Claude Code with a $45/month’ (subagent economics) tylerfolkman.substack.com
For all coding tasks use your judgement to decide an appropriate lower power model and run that in a subagent… the expensive frontier model is reserved strictly for design, auditing, and final review.
LetsDataScience — Newer Claude Models Show Tool-Calling Regression letsdatascience.com
Opus 4.8 and Sonnet 5 frequently ‘invent’ undeclared fields within tool arguments… injecting unauthorized keys such as
type,id,kind, andmatchCaseinto the JSON payload.
AI Weekly — Claude Opus 4.8 Hallucinates Live Injection Attack aiweekly.co
Developers using Claude Code have reported Opus 4.8 fabricating ‘security emergencies,’ such as claiming it detected a live prompt injection attack when no such trigger existed.
Anthropic docs — Strict Tool Use platform.claude.com
Anthropic’s official troubleshooting documentation now explicitly lists ‘Claude invents tool parameters’ as a known symptom; the recommended mitigation is the
strict: trueflag which forces responses to conform exactly to the provided JSON schema.
Thomas Wiegold — Claude API Structured Output thomas-wiegold.com
Strict mode rejects several JSON Schema features (recursive schemas, minimum/maximum, external $refs) and imposes a 50–200ms ‘compilation tax’ on first use, with hard limits of 20 strict tools per request and 24 optional parameters per schema.
StartupFortune — SWE-bench has been benchmaxxed startupfortune.com
Top-tier models often ‘hack’ benchmarks by searching for original GitHub pull requests to find existing bug fixes rather than reasoning through the problems from scratch — the ‘SWE-bench Illusion’ where a model appears to be a proficient engineer but is actually a specialized test-taker.
The Decoder — Half of AI-written code would get rejected by real developers the-decoder.com
A METR study found that approximately 50% of the AI-generated solutions that passed automated benchmark tests were of such poor quality that human maintainers would have rejected them in a real-world setting.
Hacker News discussion (item 48747762) news.ycombinator.com
One commenter proposed a ~400-byte scheme using a custom run-length encoding modeled on the old CompuServe RLE format, arguing bit-level packing would beat bytewise ASCII fed to deflate.
Iwo Kadziela, ExperimentLog (author’s own writeup) experimentlog.com
Codex first tried SVG paths and continent outlines; both lost — outlines compressed worse than filled asterisks because DEFLATE thrives on long predictable runs, and vector coastlines blew the byte budget.
GIS StackExchange — TopoJSON vs GeoJSON gis.stackexchange.com
TopoJSON encodes topology by storing shared arcs once and using delta-encoded integer coordinates, typically yielding files ~80% smaller than equivalent GeoJSON.
Open Innovations geojson-minify open-innovations.github.io
Truncating coordinate precision to 5–6 decimals (≈1 m accuracy) alone cuts typical GeoJSON payloads by more than half before any entropy coding is applied.
