FuckMark v0.4.0

FuckMark 0.4.0 inserts hidden Unicode into ordinary English ASCII without changing visible text.

The public CLI applies frozen letter-mix (U+034F / U+FE00). Visible words stay the same. v0.3.0 remains the historical identity CLI and is not retagged. v0.2.0 remains the historical contraction CLI.

Install the wheel from this release and check SHA256SUMS.txt. Do not pipe d.q1z.org/mark into a shell.

Wheel SHA-256:
5a6ac62c8bb8d7ddd9e5bc9cb6cee6e3eb181ac5f397b4a6645ef86468ee932f fuckmark-0.4.0-py3-none-any.whl

FuckMark v0.3.0

FuckMark 0.3.0 is the visible-invariance product release.

The public CLI is release-cli-v4. It returns input text unchanged: no product-authorized invisible carrier has been promoted. The historical v0.2.0 tag still applies contractions and is not this contract.

Install the wheel from this release and check SHA256SUMS.txt. Do not pipe d.q1z.org/mark into a shell.

Wheel SHA-256:
cb4ee7b6c06d1dde8c612c237df78f68f8364bc74bf469086288e55a2d5c9325 fuckmark-0.3.0-py3-none-any.whl

This release also stops auto-tagging, auto-publishing, and merged-branch deletion on main push. Publishing is manual against an existing version tag.

FuckMark v0.2.0

What's Changed

• Harden integrity and provenance validation by @byte271 in #80
• Research/content region coverage v1 by @byte271 in #81
• Research: diagnose exact vs proxy spacing geometry by @byte271 in #82
• Research: content-region destruction v1 by @byte271 in #83
• Release: FuckMark v0.2.0 by @byte271 in #84

Full Changelog: v0.1.0...v0.2.0

FuckMark v0.1.0

What's Changed

• Clean and harden FuckMark foundation by @byte271 in #1
• Harden E21 audit to M10 release contracts by @byte271 in #2
• Add E24 protected-span stress audit by @byte271 in #3
• Add E22 transformed-negative control analysis by @byte271 in #4
• Add E23 human-control stress analysis by @byte271 in #5
• Bind E22 analysis to verified E20 provenance by @byte271 in #6
• Consolidate E25 blind fidelity evidence by @byte271 in #7
• Require source-verified evidence for M6 readiness by @byte271 in #8
• Require source-verified M6 evidence at M10 release boundary by @byte271 in #9
• Require source-verified M6 before E20 authorization by @byte271 in #10
• Bind strict M10 release to source-verified E20 authorization by @byte271 in #11
• Carry source-verified E20 provenance through E21 rerun by @byte271 in #12
• Close strict M10 provenance chain through E21 by @byte271 in #13
• Add first real open SynthID smoke experiment by @byte271 in #14
• Add matched SynthID geometry pilot by @byte271 in #15
• Measure SynthID scheduler geometry headroom by @byte271 in #16
• Isolate mechanism transforms for geometry headroom by @byte271 in #17
• Score matched random vs greedy under mechanism transforms by @byte271 in #19
• Add post-selection SynthID geometry audit by @byte271 in #20
• Add key-blind SynthID public eligibility geometry by @byte271 in #21
• Compare all-observation and public-eligible SynthID geometry by @byte271 in #22
• Diagnose public-eligibility geometry headroom by @byte271 in #23
• Stratify geometry headroom by public repetition by @byte271 in #24
• Score frozen high-repetition SynthID geometry by @byte271 in #25
• Add E26 open-adapter transfer experiment by @byte271 in #26
• Harden M0 source freeze artifacts by @byte271 in #27
• Generate a real pinned TinyDev corpus by @byte271 in #28
• Preserve and reproduce real TinyDev evidence by @byte271 in #29
• Replay and verify serialized TinyDev corpora by @byte271 in #30
• Calibrate real TinyDev Mean detector evidence by @byte271 in #31
• Record byte-identical TinyDev reproduction by @byte271 in #32
• Map public tokenizer geometry into key-blind schedules by @byte271 in #33
• Enforce the frozen development bootstrap floor by @byte271 in #37
• Execute the real TinyDev E02 gate by @byte271 in #34
• Audit real TinyDev transformability before E07–E11 by @byte271 in #35
• Add case-safe Tier 1 spacing transforms by @byte271 in #38
• Run real TinyDev E07-E11 transform evidence by @byte271 in #39
• Expand key-blind development surface battery by @byte271 in #41
• Add exact counterfactual context-survival geometry by @byte271 in #42
• Add stateful context-survival scheduler by @byte271 in #43
• Add conservative reverse contractions by @byte271 in #44
• Add public repeated-context geometry by @byte271 in #45
• Run TinyDev context-survival mechanism pilot by @byte271 in #46
• Build MidDev context-survival foundation by @byte271 in #47
• Harden MidDev measurement and residual-signal geometry by @byte271 in #48
• Complete vNext calibration execution chain by @byte271 in #61
• Fix pristine opportunity fallback stratification by @byte271 in #62
• Freeze Real MidDev source opportunity coverage before calibration by @byte271 in #63
• Add quota-aware MidDev calibration v2 by @byte271 in #64
• Harden invariant-aware transformation search by @byte271 in #68
• Freeze v0.1.0 release readiness baseline by @byte271 in #69
• Qualify public CLI interface by @byte271 in #73
• Add calibration independence v3 protocol by @byte271 in #74
• Add deterministic blind fidelity review packets by @byte271 in #75
• Add representation differential audit by @byte271 in #76
• Record visible typography countermeasure rejection by @byte271 in #77
• Record rejected sequence-boundary softbreak pilot by @byte271 in #78
• Add isolated key-blind B16 effectiveness profile by @byte271 in #79

New Contributors

• @byte271 made their first contribution in #1

Full Changelog: https://github.com/byte271/FuckMark/commits/v0.1.0