oss-forensics

Investigate GitHub repositories for supply chain compromise using multi-source forensic evidence collection.

14|5|Updated Apr 9, 2026
One-click install
npx skills add https://github.com/MLT-OSS/hermes-agent-go --skill oss-forensics-mlt-oss
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: oss-forensics
Source: https://github.com/MLT-OSS/hermes-agent-go/tree/main/optional-skills/security/oss-forensics
Command: npx skills add https://github.com/MLT-OSS/hermes-agent-go --skill oss-forensics-mlt-oss

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? When a GitHub repository may have been compromised—through maintainer account takeover, dependency confusion, CI/CD injection, or force-pushed history erasure—teams lack a structured way to gather tamper-proof evidence, recover deleted commits, and produce a defensible forensic report. ## Core Features & Use Cases - 7-Phase Investigation Framework: Orchestrates IOC extraction, parallel evidence collection across five specialist investigators (local git, GitHub API, Wayback Machine, GH Archive/BigQuery, IOC enrichment), hypothesis formation, mechanical validation, and report generation. - Deleted Content Recovery: Recovers force-pushed commits via direct GitHub URLs, git fetch by SHA, git fsck dangling objects, and archived snapshots of deleted issues and PRs. - Evidence Store with Chain of Custody: A Python CLI (evidence-store.py) records every finding with SHA-256 integrity hashes, verification states, and a full chain-of-custody log. - Use Case: A maintainer suspects a contributor's account pushed a backdoored workflow file. The skill detects the force-push in GH Archive, recovers the erased commit, validates the compromise hypothesis against multi-source evidence, and outputs a structured forensic report with IOCs and mitigations. ## Quick Start Ask the agent to investigate the GitHub repository owner/repo for signs of supply chain compromise and generate a forensic report.

Frequently Asked Questions about oss-forensics

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I investigate a GitHub repository for supply chain compromise?

Run the 7-phase workflow: extract IOCs from the prompt, spawn parallel investigators for local git, GitHub API, Wayback Machine, and GH Archive, then consolidate evidence, form hypotheses, validate them mechanically, and generate a forensic report with cited evidence IDs.

How to recover force-pushed or deleted commits on GitHub?

Force-pushed commits remain on GitHub's servers until garbage collection. Recover them via the direct commit URL with a .patch suffix, the /git/commits/SHA API endpoint, git fetch origin SHA, or git fsck for dangling objects in a local clone.

Does this skill require BigQuery or Google Cloud access?

BigQuery access is optional. The GH Archive investigator requires gcloud credentials, but if unavailable the skill documents the limitation and proceeds with the other four investigators: local git, GitHub API, Wayback Machine, and IOC enrichment.

What GitHub API rate limits apply during an investigation?

Authenticated requests allow 5,000 per hour via GITHUB_TOKEN or the gh CLI, while unauthenticated requests allow only 60 per hour. The skill recommends conditional requests, sequential pagination, and pausing when X-RateLimit-Remaining drops below 100.

How does the skill prevent fabricated forensic findings?

Every claim must cite an evidence ID, and the hypothesis validator mechanically checks that each cited ID exists in the evidence store. Verified facts require confirmation from two independent sources, and hypotheses cannot be rejected without evidence-backed counter-arguments.