Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/XxYouDeaDPunKxX/canon-boundary-guard-for-gpt-project/llms.txt

Use this file to discover all available pages before exploring further.

artifact_fingerprint.py produces a JSON array of mechanical fingerprints for one or more files or directories. For each path it emits the path as given on the command line, an existence flag, file/directory type, size in bytes, last-modified time in UTC ISO-8601 format, and a SHA-256 hex digest. Fingerprinting is the recommended first step when a bundle zip arrives in /mnt/data — recording the hash before extraction anchors the source identity required for treating extracted surfaces as L0. The script does not classify provenance; it only records facts about the file system.

​
Usage

# Fingerprint a single file
python scripts/artifact_fingerprint.py /path/to/file.json

# Fingerprint multiple files and a directory
python scripts/artifact_fingerprint.py /path/to/file1 /path/to/file2 /path/to/dir

# Fingerprint a zip with 4-space JSON indentation
python scripts/artifact_fingerprint.py /path/to/file.zip --indent 4

​
Arguments

​
paths
Path (one or more, positional)
One or more files or directories to fingerprint. Each path is processed independently. Directories receive size and mtime but no SHA-256 (hashing is file-only).
​
--indent
int
default:"2"
JSON indentation level for the output array. Defaults to 2. Pass 0 for compact single-line output.

​
Output

The script prints a single JSON array to stdout. Each element corresponds to one input path. When the path exists and is a file: 
[
  {
    "path": "/mnt/data/canon-boundary-guard-gpt.zip",
    "exists": true,
    "is_file": true,
    "is_dir": false,
    "size_bytes": 14823,
    "mtime_utc": "2025-05-23T17:38:50+00:00",
    "sha256": "a3f5..."
  }
]
When the path does not exist, only path and exists: false are included: 
[
  {
    "path": "/mnt/data/missing.zip",
    "exists": false
  }
]
When the path is a directory, sha256 is omitted; all other fields are present.

​
Field reference

FieldTypeDescription
pathstringThe path as passed on the command line
existsbooleanWhether the path exists on the file system
is_filebooleantrue when the path is a regular file
is_dirbooleantrue when the path is a directory
size_bytesintegerFile or directory size from stat
mtime_utcstringLast-modified time as UTC ISO-8601 with timezone offset
sha256stringFull SHA-256 hex digest, present for files only

​
How it fits the protocol

The sha256 value serves as the source identity anchor required when treating source-staged extractions as L0. Before a bundle zip is extracted, the protocol requires recording the source zip path and hash — if no anchor is available, L0 must be limited to the inspected path with a risk note. Running artifact_fingerprint.py on the zip before extraction produces a machine-readable record that satisfies this requirement. Store the output alongside SESSION_STATE or include the hash in active_l0_sources when initialising state.
Run this script on the bundle zip before extracting it, to record the source anchor in SESSION_STATE. A fingerprint taken after extraction still identifies the zip, but recording it first removes any ambiguity about which file was the extraction source.

Build docs developers (and LLMs) love

Get started for freeTalk to us