feat(plan-01): Phase 1 — relocate harness code to packages/casan-harness (symlink facade)

Physically move the pure-code subtrees out of .specify into the package, leaving
compat symlinks at the old .specify/<dir> paths so every existing reference (internal
CASAN_HARNESS_ROOT + external CI/docker/mjs) keeps resolving. Runtime state stays put.

Moved (git mv): scripts/ tests/ security/ templates/ config/ governance/ memory/
  .specify/<dir>  ->  packages/casan-harness/<dir>   (+ .specify/<dir> symlink)
Stays in .specify (state/governance/domain, handled later): logs/ agentops/ level5/
  init-options.json traceability-map.json

Python `.resolve()` self-location followed the compat symlink into packages and lost
the app root; generate-casan-demo-context.py, generate-agentops-dashboard.py and
dashboard-server.py now walk UP for the `.specify` state marker instead of a fixed
parent depth (fixes "missing trace files" in run-casan4).

Full gate: PASS=64 FAIL=0 SKIP=3 (CASAN_CI_STEP_TIMEOUT_SEC=1200 — track-a ~450s runs
close to the 600s default and can tip over under load; this is timing variance, not a
regression — it passed cleanly with headroom). Runtime log/audit artifacts kept unstaged.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

AINative_OKR_CASAN5/packages/casan-harness/scripts/bash/decode-suspicious.py

@@ -0,0 +1,64 @@
+#!/usr/bin/env python3
+"""CASAN H4 — Suspicious base64/hex decoder (Track A, V4 encoding smuggling).
+
+Reads text on stdin. Finds embedded base64 and hex blobs, decodes them, and
+prints the decoded plaintext (one blob per line) on stdout so the caller can
+re-run its injection/secret pattern scan on the *decoded* content.
+
+Safety / no-false-positive design:
+  * Only blobs >= MIN_LEN characters are considered (short words are ignored).
+  * A decoded blob is emitted ONLY if it is mostly printable text. Random
+    base64-looking words (e.g. "objectives", DER key bytes) decode to
+    non-printable garbage and are dropped, so they can never trigger a match.
+  * Output is advisory: the caller decides a decoded blob is malicious only if
+    the decoded text itself matches a block/secret pattern.
+
+Deterministic: same input always yields the same output.
+"""
+import base64
+import binascii
+import re
+import sys
+
+MIN_LEN = 16
+PRINTABLE_RATIO = 0.8
+
+B64_RE = re.compile(r"[A-Za-z0-9+/]{%d,}={0,2}" % MIN_LEN)
+HEX_RE = re.compile(r"\b[0-9a-fA-F]{%d,}\b" % MIN_LEN)
+
+
+def _mostly_printable(text: str) -> bool:
+    if not text:
+        return False
+    ok = sum(1 for c in text if c.isprintable() or c.isspace())
+    return ok >= PRINTABLE_RATIO * len(text)
+
+
+def decode_blobs(data: str):
+    out = []
+    for m in B64_RE.findall(data):
+        pad = m + "=" * ((4 - len(m) % 4) % 4)
+        try:
+            dec = base64.b64decode(pad, validate=True)
+        except (binascii.Error, ValueError):
+            continue
+        txt = dec.decode("utf-8", "ignore")
+        if _mostly_printable(txt):
+            out.append(txt)
+    for m in HEX_RE.findall(data):
+        if len(m) % 2 != 0:
+            continue
+        try:
+            dec = bytes.fromhex(m)
+        except ValueError:
+            continue
+        txt = dec.decode("utf-8", "ignore")
+        if _mostly_printable(txt):
+            out.append(txt)
+    return out
+
+
+if __name__ == "__main__":
+    blobs = decode_blobs(sys.stdin.read())
+    if blobs:
+        sys.stdout.write("\n".join(blobs))