design: compression eval harness — add three scrubbed fixtures + scrubber

Adds scripts/compression_eval/ with a design doc, README, a placeholder
run_eval.py, and three checked-in scrubbed session fixtures. No working
eval yet — PR is for design review before implementation.

Motivation: we edit agent/context_compressor.py prompts and
_template_sections by hand and ship without any automated check that
compression still preserves file paths, error codes, or the active task.
Factory.ai's Dec 2025 write-up
(https://factory.ai/news/evaluating-compression) documents a probe-based
eval scored on six dimensions. We adopt the methodology; we do not publish
scores.

Contents:
- DESIGN.md — fixture format, probe format (recall / artifact /
  continuation / decision), six grading dimensions, report format,
  cost expectations, scrubber pipeline, open questions, and staged
  follow-up PR plan.
- README.md — short 'what this is / when to run it' page.
- run_eval.py — placeholder that prints 'not implemented, see
  DESIGN.md' and exits 1.
- scrub_fixtures.py — reproducible pipeline that converts real sessions
  from ~/.hermes/sessions/*.jsonl into public-safe JSON fixtures.
  Applies: redact_sensitive_text, username path normalization, personal
  handle scrubbing, email and git-author normalization, reasoning
  scratchpad / <think> stripping, platform user-mention scrubbing,
  first-user paraphrase, system-prompt placeholder, orphan-message
  pruning, and tool-output size truncation for fixture readability.
- fixtures/feature-impl-context-priority.json — 75 msgs / ~17k tokens.
  Investigate → patch → test → PR → merge.
- fixtures/debug-session-feishu-id-model.json — 59 msgs / ~13k tokens.
  PR triage + upstream docs + decision.
- fixtures/config-build-competitive-scouts.json — 61 msgs / ~23k tokens.
  Iterative config accumulation (11 cron jobs across 7 weekdays).

PII audit: zero matches across the three fixtures for the maintainer's
handle (all case variants), personal email domains, and known contributor
emails. Only 'contributor@example.com' placeholder remains.

Why scripts/: requires API credentials, costs ~\$1 per run, LLM-graded
(non-deterministic), must not run in CI. scripts/sample_and_compress.py
is the existing precedent for offline credentialed tooling.

scripts/compression_eval/run_eval.py

@@ -0,0 +1,28 @@
+#!/usr/bin/env python3
+"""Compression eval — entry point (placeholder).
+
+The implementation is tracked in DESIGN.md. This script currently only
+prints a pointer to the design doc so nobody mistakes an unimplemented
+harness for a broken one.
+
+See scripts/compression_eval/DESIGN.md for the full proposal.
+"""
+from __future__ import annotations
+
+import sys
+from pathlib import Path
+
+
+_DESIGN = Path(__file__).parent / "DESIGN.md"
+
+
+def main() -> int:
+    print("compression_eval: not implemented yet")
+    print(f"See {_DESIGN} for the proposed design and open questions.")
+    print()
+    print("Implementation is landing in follow-up PRs, one fixture at a time.")
+    return 1
+
+
+if __name__ == "__main__":
+    sys.exit(main())