bootloader-uboot

U-Boot bootloader attack playbook — console interrupt to break the autoboot countdown, environment variable inspection and manipulation, bootargs override to spawn init=/bin/sh, secure-boot bypass techniques, and fault-injection basics (voltage and clock glitching). Covers MIPS, ARM, and AArch64 targets.

Quality

73%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./packages/decepticon/decepticon/skills/standard/iot/bootloader-uboot/SKILL.md

Quality

Content

65%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A highly actionable, well-sequenced U-Boot attack playbook that earns top marks for concrete commands and code. It is weakened by verbosity, the lack of explicit validation feedback loops around destructive operations, and a monolithic structure that puts large scripts inline instead of in reference files.

Suggestions

Add explicit validation checkpoints to destructive operations (e.g. verify mtdparts before `sf write`, confirm `=>` prompt before `run bootcmd`) as validate→fix→retry loops to lift workflow clarity.

Move the full automation Python script and the ChipWhisperer glitching code into `scripts/` reference files, linking to them from the body to enable progressive disclosure and trim the inline token budget.

Cut explanatory prose that restates concepts Claude already knows (what U-Boot is, what fault injection does) and keep only command-specific notes to improve conciseness.

Dimension	Reasoning	Score
Conciseness	Mostly actionable but padded with explanatory prose Claude largely knows (e.g. "replace the kernel init process with a root shell before any authentication or privilege separation occurs") and a ~330-line body that could be tightened; not level 3 because several passages restate embedded-boot concepts, not level 1 because the bulk is concrete commands rather than concept exposition.	2 / 3
Actionability	Provides fully executable commands and code — `setenv bootargs`, `printenv`, `sf write`, and a complete ChipWhisperer Python script — that are copy-paste ready; not level 2 because examples are concrete and complete rather than pseudocode or missing key details.	3 / 3
Workflow Clarity	Phases 1–6 give a clear sequence but destructive operations (saveenv, sf write, voltage glitching) carry warnings rather than explicit validate→fix→retry checkpoints, so per the feedback-loop guidance this is capped at 2; not level 1 because steps are sequenced, not level 3 because validation checkpoints are missing.	2 / 3
Progressive Disclosure	Well-organized into phases but monolithic — no bundle files exist and large content that should be separate (the full automation Python script, the ChipWhisperer glitching code) is inline with only external URLs for navigation; not level 3 because it is far over 50 lines with no one-level-deep file references, not level 1 because sections are clearly structured rather than a disorganized wall.	2 / 3
	Total	9 / 12 Passed

Description

82%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A specific, well-scoped description with strong concrete actions and natural trigger terms. Its main weakness is the absence of an explicit "Use when..." clause, leaving the "when to use" guidance implied rather than stated.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "console interrupt to break the autoboot countdown", "environment variable inspection and manipulation", "bootargs override to spawn init=/bin/sh", "secure-boot bypass techniques", "fault-injection basics (voltage and clock glitching)" — matching the level-3 anchor of several specific concrete actions; not level 2 because coverage is comprehensive rather than partial.	3 / 3
Completeness	Strongly answers "what" with enumerated actions but has no explicit "Use when..." trigger clause, only an implied scope, so per the guideline a missing explicit trigger caps this at 2; not level 3 because "when" is never stated and not level 1 because "what" is detailed.	2 / 3
Trigger Term Quality	Contains natural terms a user would say — "U-Boot", "bootloader", "autoboot", "bootargs", "secure boot bypass", "fault injection", "glitching" — giving good coverage of common variations; not level 2 because it spans both the generic and the specific vocabulary.	3 / 3
Distinctiveness Conflict Risk	A clear niche (U-Boot bootloader attacks on embedded targets) with distinctive triggers unlikely to fire for other skills; not level 2 because the domain is narrow and well-scoped, not generic.	3 / 3
	Total	11 / 12 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning

	Total	15 / 16 Passed

Repository: PurpleAILAB/Decepticon
Path: packages/decepticon/decepticon/skills/standard/iot/bootloader-uboot/SKILL.md
Commit: e34afba

Reviewed: about 15 hours ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.