Anti-Forensics Field Manual

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
====== Chain of Custody ======

===== Definition =====
Chain of Custody is a core forensic concept used to keep digital evidence understandable, defensible, and tied to the limits of the collection.
In a forensic report, chain of custody should be treated as a source of evidence and uncertainty, not as a shortcut to intent. The useful question is what the record supports, what it does not support, and what another source says when asked the same question.

===== Background =====
The concept matters because investigations are built from small technical claims. One weak claim can drag a clean report into unallocated space and make the reviewer do excavation work.
Good analysis keeps the artifact in its lane: useful, limited, and not allowed to perform intent for the room. A neat conclusion without notes is just a polished fragment looking for a file header.

===== Technical Description =====
In practice, chain of custody depends on source identification, collection conditions, record integrity, time handling, and the difference between an observed fact and an interpretation.
Version, platform, retention, and collection scope should be documented before the artifact is promoted to a finding.

===== Forensic Relevance =====
  * **Contradiction testing:** It is useful for finding places where logs, metadata, storage behavior, or accounts disagree.
  * **Reporting decisions:** It helps decide how narrow the finding should be when chain of custody looks interesting but not conclusive.
  * **Timeline reconstruction:** Chain of Custody can help place activity in sequence when the time source and collection conditions are understood.
  * **Recovery analysis:** It can explain why data was recovered, missed, corrupted, or only partially reconstructed.
  * **User activity review:** It may support account or profile context, but only when independent artifacts point the same direction.

===== Evidence Sources =====
^ Evidence source ^ What it may show ^ Reliability limits ^ What it cannot prove alone ^
| Case notes | Show what was collected, when it was collected, and which assumptions were made. | Can be incomplete, rushed, or written after the examiner already formed a theory. | Do not prove the underlying event by themselves. |
| Acquisition logs | Record tools, hashes, source identifiers, and collection errors. | Depend on accurate tool configuration and disciplined note-taking. | Do not explain user intent. |
| Hash records | Support integrity comparison between originals, images, and working copies. | Only cover the data actually hashed. | Do not prove that the content is meaningful to the case. |
| Peer review notes | Show whether a second reviewer could follow the same reasoning. | Can miss errors if the reviewer only checks formatting. | Do not replace artifact-level corroboration. |

===== Interpretation Limits =====
The most common error is treating chain of custody as proof of motive. It may support a technical event, a sequence, or a contradiction, but motive needs stronger ground.
Another bad leap is treating absence as intent. Missing data can come from retention, configuration, collection scope, sync behavior, overwriting, media behavior, or ordinary use.
A tool result should be described as a parsed or recovered record, not as the tool's opinion about what happened. Tools surface evidence; they do not understand it.
The evidence does not become stronger because the report sounds certain; it just becomes harder to clean up.

===== Common Misinterpretations =====
  * Ignoring benign explanations because the suspicious explanation is easier to write.
  * Treating chain of custody as proof that a specific user acted, when the artifact only supports system or account context.
  * Treating a timestamp as exact truth without checking timezone, clock drift, and source semantics.

===== Example Scenario =====
A reviewer sees case notes in the case file and asks whether it actually supports the written conclusion.
The artifact may explain part of the sequence, but ordinary system behavior still needs to be ruled in or out.
The careful next step is to normalize the time source, compare independent artifacts, and write the finding as support rather than proof.

===== Analysis Workflow =====
  - Define the question before opening another parser: what should Chain of Custody help answer?
  - Preserve the source evidence and document how the case notes were collected.
  - Record tool versions, input paths, output paths, time settings, and errors.
  - Identify observed facts before writing any interpretation.
  - Normalize time sources and document timezone, clock drift, and collection-time effects.
  - Compare at least two independent artifact families before raising confidence.
  - Consider benign explanations, automated behavior, retention, sync, and storage-device behavior.
  - Write conclusions proportionally: observed fact first, inference second, uncertainty always visible.

===== Reporting Guidance =====
Reporting on chain of custody should be precise enough that another analyst can retrace the claim without inheriting the original examiner's confidence.
Avoid wording that converts possibility into intent. The report should not say a user deliberately deleted, hid, wiped, or tampered with evidence unless the evidence actually supports that conclusion.
Report-ready wording:
  * The available case notes are consistent with activity related to chain of custody, but they do not independently establish motive or user intent.
  * Recovery was limited under the examined conditions; additional corroboration would be required before concluding deliberate destruction.
  * The finding should be read with the collection scope, time-source limits, and alternate explanations described in this report.

===== Confidence and Reliability =====
^ Confidence level ^ What it looks like for this topic ^ How to report it ^
| Low | Case notes exist, but collection scope, time source, or surrounding context is limited. | State the observation and keep interpretation narrow. |
| Moderate | Case notes align with Acquisition logs, but attribution or intent remains unresolved. | Say the artifacts support the finding, not that they prove it. |
| High | Multiple independent sources agree on sequence, source system, account context, and collection conditions. | Use stronger language, but still separate observed facts from inference. |

===== Related Concepts =====
  * [[wiki:digital_forensics:foundations:acquisition_notes|Acquisition Notes]] - Often appears nearby in reviews of the same system or behavior.
  * [[wiki:digital_forensics:foundations:artifact_corroboration|Artifact Corroboration]] - Often appears nearby in reviews of the same system or behavior.
  * [[wiki:digital_forensics:foundations:case_chronology|Case Chronology]] - Useful when checking whether another artifact supports the same interpretation.
  * [[wiki:digital_forensics:foundations:case_notes|Case Notes]] - Helps keep the finding grounded in a wider evidence pattern.
  * [[wiki:digital_forensics:foundations:collection_bias|Collection Bias]] - Compare this concept when testing sequence and corroboration.
  * [[wiki:digital_forensics:foundations:dead_box_analysis|Dead-Box Analysis]] - Helps keep the finding grounded in a wider evidence pattern.

===== Tools =====
  * [[https://www.autopsy.com/|Autopsy]] - Case review, file-system browsing, timelines, term search, and recovered artifact review.
  * [[https://www.sleuthkit.org/sleuthkit/|The Sleuth Kit]] - Command-line file-system analysis and evidence inspection.
  * [[https://www.exterro.com/ftk-product-downloads|FTK Imager]] - Forensic imaging, preview, and export workflows.
  * [[https://www.magnetforensics.com/products/magnet-axiom/|Magnet AXIOM]] - Commercial forensic suite for computer, mobile, and cloud evidence review.
  * [[https://github.com/log2timeline/plaso|Plaso]] - Timeline extraction and normalization across many artifact types.
  * [[https://github.com/google/timesketch|Timesketch]] - Collaborative timeline analysis and event tagging.
  * [[https://guymager.sourceforge.io/|Guymager]] - Linux forensic imaging with hashing and acquisition logs.
  * [[https://sourceforge.net/projects/dc3dd/|dc3dd]] - Disk imaging and hashing utility.
  * [[https://www.gnu.org/software/ddrescue/|GNU ddrescue]] - Imaging damaged media while preserving recovery logs.
  * [[https://arsenalrecon.com/products/arsenal-image-mounter|Arsenal Image Mounter]] - Mount forensic images for controlled review.
  * [[wiki:tools:start|Forensic Tools]] - Full tool directory for the wiki.

===== Limitations of Tools =====
Tools parse, surface, and organize evidence. They do not create conclusions.
Parser output can be affected by version differences, unsupported formats, corrupted records, timezone handling, partial collection, and storage behavior outside the tool's view.
When a tool produces a strong-looking result, validate it against another tool or source where the stakes justify it. A parser can recover a fragment; it cannot tell you whether the fragment deserves a paragraph in the report.

===== References and Further Reading =====
  * [[https://csrc.nist.gov/publications/detail/sp/800-86/final|NIST SP 800-86]] - Guide to integrating forensic techniques into incident response.
  * [[https://www.nist.gov/itl/ssd/software-quality-group/computer-forensics-tool-testing-program-cftt|NIST CFTT]] - Computer Forensics Tool Testing program.
  * [[https://www.forensicfocus.com/articles/what-is-digital-forensics/|Forensic Focus: What is digital forensics?]] - Practitioner-facing overview.

===== See Also =====
  * [[wiki:digital_forensics:foundations:start|Foundations]]
  * [[wiki:digital_forensics:foundations:acquisition_notes|Acquisition Notes]]
  * [[wiki:digital_forensics:foundations:artifact_corroboration|Artifact Corroboration]]
  * [[wiki:digital_forensics:foundations:case_chronology|Case Chronology]]
  * [[wiki:digital_forensics:foundations:case_notes|Case Notes]]
  * [[wiki:digital_forensics:foundations:collection_bias|Collection Bias]]
  * [[wiki:digital_forensics:foundations:dead_box_analysis|Dead-Box Analysis]]
  * [[wiki:digital_forensics:foundations:digital_evidence|Digital Evidence]]
  * [[wiki:digital_forensics:foundations:digital_forensics|Digital Forensics]]
  * [[wiki:digital_forensics:foundations:evidence_integrity|Evidence Integrity]]
  * [[wiki:digital_forensics:foundations:evidence_packaging|Evidence Packaging]]
  * [[wiki:tools:start|Forensic Tools]]
  * [[wiki:resources:start|Project Resources]]

===== Reader Takeaway =====
Chain of Custody is useful when it helps explain what the evidence can support and where the limits begin. Treat chain of custody as one part of a corroborated record, not a shortcut to intent. The evidence does not become stronger because the report sounds certain; it just becomes harder to clean up.

===== Use Notes =====
This article is for defensive education and technical reference. It should not be treated as legal, forensic, investigative, compliance, or operational advice without qualified professional judgment.