Works with the AI Agents you already use
FirmGraph Firmware Checks Demo.
Run a local FirmGraph demo to detect and verify firmware configuration vulnerabilities and hardcoded secrets.
Free
FirmGraph Firmware Checks Demo.
Example session with this skill installed
Run the FirmGraph demo in ./demo-work and ./demo-out to show me how it detects hardcoded secrets in firmware configurations.
- Read your context and instructions
- Compiled the firmgraph firmware checks
- Generated the document
The FirmGraph ChecksEngine has successfully processed the synthetic fixtures. The initial scan identified five security violations, which were subsequently remediated and verified in the second pass.
FirmGraph Demo Security Findings
| Finding ID | Severity | Category | Evidence (Redacted) | Status |
|---|---|---|---|---|
| FG-SEC-001 | Critical | Hardcoded Key | [REDACTED_HASH:8f3e...] | Fixed |
| FG-SEC-002 | Critical | Hardcoded Password | [REDACTED_HASH:a1b2...] | Fixed |
| FG-SET-001 | High | Insecure Protocol | telnet=enabled | Fixed |
| FG-SET-002 | High | Root Login | permit_root_login=yes | Fixed |
| FG-SET-003 | High | Weak Encryption | cipher=des | Fixed |
Execution Manifest Summary
- Engine Version:
8bde16f - Baseline Findings: 5
- Verification Findings: 0
- Artifacts: Generated
report.md,findings.json, andmanifest.jsonin./demo-out.
Changes Applied for Verification
- Replaced the RSA private key header with a reference to a secure keystore.
- Updated the configuration fixture to disable Telnet and permit only SSH with public-key authentication.
- Upgraded the cipher suite from DES to AES-256-GCM.
Next steps
- Open
./demo-out/report.htmlin your browser to view the formatted evidence. - Review the
manifest.jsonfile for the full SHA-256 audit trail of the fixtures. - Compare the
beforeandafterfixtures in the workspace to see the specific remediation patterns.
report.html
PDF · document
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Manual firmware audits are slow and prone to missing hardcoded secrets or insecure configurations. This skill provides a controlled environment to run the open-source FirmGraph ChecksEngine against synthetic fixtures, demonstrating how automated pattern matching identifies vulnerabilities in firmware files.
What it does
- Environment setup initializes a pinned version of the FirmGraph engine and a dedicated virtual environment.
- Pattern detection identifies five specific security findings including hardcoded secrets and insecure configurations.
- Remediation verification generates a corrected fixture and reruns checks to confirm a zero-finding state.
- Evidence generation produces JSON findings, SHA-256 hashes, and local HTML/Markdown reports.
- Redaction management handles sensitive evidence strings using hashing and placeholder values.
How it works
- System Check: Runs the
doctorcommand to verify Python 3.12+, Git, anduvare available. - Engine Setup: Downloads and installs the FirmGraph engine at a specific commit into a local workspace.
- Demo Execution: Processes synthetic fixtures to produce five baseline security findings.
- Verification: Applies configuration changes and verifies the remediated files pass all checks.
Frameworks & tools
This skill uses Python 3.12, the FirmGraph ChecksEngine, and uv for package management. It operates entirely on the local filesystem and requires terminal access.
Why this beats prompting it yourself
Prompting an LLM to "find bugs" often results in hallucinations or missed edge cases in configuration files. This skill executes a specific, deterministic engine that uses formal rules to ensure reproducible results and verifiable evidence.
Use cases
- Tool evaluation by testing FirmGraph's detection logic before integration.
- Security training to show developers how hardcoded secrets appear in firmware.
- Pipeline prototyping for automated firmware configuration checking.
Known limitations
Does not perform binary analysis, CVE matching, or SBOM creation. Tested on macOS; Windows is not supported. Requires a local terminal and Python 3.12.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Security.
Requires Python 3.12+ and terminal access. Does not perform binary analysis, CVE matching, or SBOM creation. Tested on macOS; Windows is not supported.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 9 days ago
- Free with an Agensi account
- Software
Python