low-level-dev-skills
AGENTS.md290 starsChanged 7 months agoMITView on GitHub
Guidelines for AI agents working in this repository. A curated suite of AI agent skills for systems and low-level C/C++ development toolchains. Each skill is a markdown file following the SKILL.md format. Skills are organized by category and cross-reference each other. Each skill has a SKILL.md file and optional references/ for detailed content. Every SKILL.md must have: bash # Comment explains purpose gcc -Wall -Wextra -O2 -o prog main.c Question? ├── Yes → Action A └── No → Action B
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 31,439 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository142 other files it gives its agents.
- custom-allocatorsSkill
- numa-programmingSkill
- af-xdpSkill
- dpdkSkill
- io-uringSkill
- adc-dac-baremetalSkill
- baremetal-startupSkill
- bootloaders-embeddedSkill
- datasheet-and-refmanual-readingSkill
- dma-baremetalSkill
- gpio-baremetalSkill
- interrupts-and-exceptions-baremetalSkill
- low-power-embeddedSkill
- mmio-and-bit-manipulationSkill
- peripherals-from-datasheetSkill
- spi-i2c-baremetalSkill
- stm32-baremetalSkill
- timers-pwm-baremetalSkill
- uart-serial-baremetalSkill
- binutilsSkill
- dynamic-linkingSkill
- elf-inspectionSkill
- linkers-ltoSkill
- bazelSkill
- build-accelerationSkill
- cmakeSkill
- conan-vcpkgSkill
- include-what-you-useSkill
- makeSkill
- mesonSkill
- ninjaSkill
- static-analysisSkill
- code-generation-and-backendsSkill
- compiler-frontendSkill
- compiler-optimizations-deepSkill
- jit-compilationSkill
- llvm-ir-and-passesSkill
- llvm-passesSkill
- mlirSkill
- clangSkill
- cpp-modulesSkill
- cpp-templatesSkill
- cross-gccSkill
- gccSkill
- llvmSkill
- msvc-clSkill
- pgoSkill
- abi-and-calling-conventionsSkill
- branch-prediction-and-speculationSkill
- cpu-pipelines-and-hazardsSkill
- memory-hierarchy-and-cachesSkill
- virtual-memory-paging-and-tlbSkill
- concurrency-debuggingSkill
- core-dumpsSkill
- debug-optimized-buildsSkill
- dwarf-debug-formatSkill
- gdbSkill
- lldbSkill
- embedded-rustSkill
- freertosSkill
- The first 60 shown.
The file
We split this AGENTS.md into the rules it gives an agent
10 rules. About 0 are common, 7 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 59
Every SKILL.md must have:
- Distinctive: line 81
Use backticks for skill references:
- Distinctive: line 131
Add required sections (Purpose, Triggers, Workflow, Related skills)
- Very specific: line 132
Create references/ if content exceeds 500 lines
- Distinctive: line 133
Update README.md skills table
- Distinctive: line 134
Add cross-references from 2+ related skills
- Distinctive: line 138
Keep body under 500 lines
- Very specific: line 146
Verify frontmatter has name and description
- Distinctive: line 147
Check all code blocks have language tags
- Distinctive: line 149
Ensure Related skills section has 2+ entries
≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
AGENTS.md
Guidelines for AI agents working in this repository.
Project Overview
A curated suite of AI agent skills for systems and low-level C/C++ development toolchains. Each skill is a markdown file following the SKILL.md format. Skills are organized by category and cross-reference each other.
Repository Structure
skills/
├── compilers/ # gcc, clang, llvm, msvc-cl, cross-gcc
├── debuggers/ # gdb, lldb, core-dumps
├── profilers/ # linux-perf, valgrind, flamegraphs
├── build-systems/ # cmake, make, ninja
├── binaries/ # elf-inspection, linkers-lto, binutils
├── runtimes/ # sanitizers, fuzzing
└── low-level-programming/ # assembly-x86, assembly-arm, interpretersEach skill has a SKILL.md file and optional references/ for detailed content.
Validation Commands
# Validate markdown formatting (requires markdownlint-cli)
markdownlint skills/**/*.md
# Check skill frontmatter is valid YAML
head -5 skills/compilers/gcc/SKILL.md
# Verify all skills have required sections
grep -l "## Purpose" skills/**/SKILL.md | wc -l
grep -l "## Triggers" skills/**/SKILL.md | wc -l
grep -l "## Workflow" skills/**/SKILL.md | wc -l
grep -l "## Related" skills/**/SKILL.md | wc -l
# Check cross-reference format (should use backticks, not links)
grep -r "skills/[a-z]" skills/ --include="*.md" | grep -v "`skills/"SKILL.md Format
YAML Frontmatter
---
name: skill-name
description: <skill-name> skill for <domain>. Use when <scenarios>. Activates on <triggers>.
---name: Lowercase, hyphen-separated, matches directory namedescription: Must include "Use when" and "Activates on" clauses
Required Sections
Every SKILL.md must have:
- Purpose (1-2 sentences): What the skill does
- Triggers (4-6 items): User query patterns that activate this skill
- Workflow (numbered subsections): Step-by-step guidance with:
- Decision trees (ASCII in code blocks)
- Command tables
- Code examples with language tags
- Error/symptom tables
- Related skills (2+ references): Cross-references using backtick format
Code Style Guidelines
Markdown
- Maximum body: 500 lines (put exhaustive lists in
references/) - Code blocks: Always include language tag (
```bash,```c,```cmake) - Tables: Pipe syntax with header separator
- Lists:
-for unordered,1.for ordered
Cross-References
Use backticks for skill references:
- Correct:
skills/compilers/gcc - Incorrect: "the gcc skill" or
[gcc](skills/compilers/gcc)
For internal references: [references/flags.md](references/flags.md)
Naming Conventions
- Skill directories: lowercase, hyphen-separated (
linux-perf,cross-gcc) - SKILL.md: always uppercase
- Reference files: lowercase with extension (
flags.md,cheatsheet.md)
Code Examples
Code examples should be:
- Complete and compilable where possible
- Annotated with comments explaining purpose
- Grouped with blank lines between related commands
# Comment explains purpose
gcc -Wall -Wextra -O2 -o prog main.c
# Related command follows blank line
./progTables
| Command | Effect |
cmd | Description |
| Symptom | Cause | Fix |
| Error message | Root cause | Resolution |
Decision Trees
Question?
├── Yes → Action A
└── No → Action BCommon Workflows
Adding a New Skill
- Create
skills/<category>/<skill-name>/SKILL.md - Add required sections (Purpose, Triggers, Workflow, Related skills)
- Create
references/if content exceeds 500 lines - Update README.md skills table
- Add cross-references from 2+ related skills
Updating an Existing Skill
- Keep body under 500 lines
- Move exhaustive flag lists to
references/ - Maintain numbered workflow subsections
- Preserve existing cross-references
Validating Changes
- Run
markdownlint skills/**/*.md - Verify frontmatter has
nameanddescription - Check all code blocks have language tags
- Confirm cross-references use backtick format
- Ensure Related skills section has 2+ entries
Quality Checklist
- Frontmatter:
nameanddescriptionfields present - Description: includes "Use when" and "Activates on"
- Purpose: 1-2 sentences
- Triggers: 4-6 user query patterns
- Workflow: numbered subsections (1, 2, 3...)
- Code blocks: language tags included
- Related skills: 2+ references in backtick format
- Body length: under 500 lines
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

