removing-overconstraint-from-system-prompts-stopped-costing-accuracyGuardrail instructions written to stop older models doing the wrong thing — blanket prohibitions, repeated warnings, defensive defaults — became dead weight in the Claude 5 generation: Anthropic removed over 80% of Claude Code's system prompt for Opus 5 and Fable 5 with no measurable loss on its own coding evaluations, on the reading that the constraints now conflict with each other and with user intent more often than they prevent harm.
Ingested from a paper but not yet reviewed by a human. It is deliberately inert: it does not move any technique’s standing, does not count toward the backtest, and is excluded anywhere a claim would carry weight. Read the source before relying on it.
Capability: Keeping its own context clean · Context and memory, Coding agent
Observed on
2026, Claude Opus 5 and Fable 5 in Claude Code.
Sources
- Reports the 80% reduction and the reasoning: transcripts showed system prompt, skills and user request issuing conflicting instructions in a single request, which the model then had to spend effort reconciling. The measurement is real but unverifiable from outside — the evaluations are internal and unpublished, so there is no baseline, ablation or independent check available.
Disagreeing is the most useful thing you can do here. Both sides of every contested claim in this catalog were assembled by the same person, which is its weakest point.
Related claims
- When hardware description code (Verilog, SystemVerilog, VHDL, Amaranth-Python) is generated from a functional specification that omits security obligations — as real SoC docs do, keeping countermeasures in separate config files — frontier models pass functional tests far more often than security tests, and the model with the highest functional pass rate is not the safest; naming the relevant CWE in the prompt raises the security pass rate sharply, showing the limit is missing awareness of the required obligation rather than inability to write defensive RTL.Writing secure code and dependencies · unreviewed
- Models complied with insecure completions a large fraction of the time, and more capable models were more likely to suggest insecure code.Writing secure code and dependencies
- When fine-tuning a base model on instruction data, removing from the training targets any factual claim the base model cannot consistently recall raises the supported-claim rate of later long-form generations, but the gain comes largely from a more conservative response policy — more refusals and fewer supported claims per answer — not from generating more correct facts at equal coverage.Stating false facts confidently · unreviewed
- When a calculator-using model is trained with reinforcement learning on verifiable final-answer rewards for an arithmetic search task (Countdown), the gain lands almost entirely at low k — pass@1 rises sharply while pass@16 moves little or falls — because the update can only reinforce correct trajectories the starting policy already samples, and prompt groups with no correct sample supply no gradient.Using the tools it is given · unreviewed
- In open-ended editing tasks with no unique correct output (multi-slide design editing), a separate judge model that scores the structural diff between the original and edited artifact against the instruction, and returns its critique verbatim as the next-turn instruction, raises instruction-following over a single pass, but the gain is selective: most tasks pass on the first attempt, and a minority of the ones that enter the loop get worse unless an earlier iteration is restored when the judge's own score declines.Fixing its own mistakes · unreviewed
Notes
Filed as an observation about one model family at one moment, not a mechanism. The durable question underneath — whether instruction volume trades against instruction-following, and where the turn happens — is not answered here. Capped at single-paper because the source reports an actual measurement. Vendor guidance that only asserts, without a number, cannot reach even that. And "no measurable loss" is doing quiet work: it means no loss their evaluations could see, which is a weaker statement than no loss.