hashintel/skeleton
pocs/agent-plugins/claude-pragma/skills/skeleton/SKILL.md
Build a walking skeleton — the thinnest runnable system that proves build, test, and runtime work end-to-end. Use when starting a new project, before any feature work. Front-loads tooling and infrastructure so every subsequent slice is cheaper.
$ install --global
skillsauthnpx skillsauth add hashintel/labs skeletonInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: May 15, 2026, 3:11 AM258.8s1 file scanned
SKILL.md
- name:
- skeleton
- description:
- Build a walking skeleton — the thinnest runnable system that proves build, test, and runtime work end-to-end. Use when starting a new project, before any feature work. Front-loads tooling and infrastructure so every subsequent slice is cheaper.
- argument-hint:
- [project description and tech stack]
Walking Skeleton
Build a walking skeleton (Cockburn) — the thinnest possible runnable system that proves the entire pipeline works: build, test, run.
Input
The project description and tech stack: $ARGUMENTS
Precondition: a concept capsule must exist. If none exists, stop and recommend /pragma:capsule first. Use glossary terms even for the "hello world" path.
What a skeleton IS and IS NOT
IS: project builds, test runner runs with ≥1 passing test, one minimal path executes end-to-end, happy-path scenario from capsule has a placeholder.
IS NOT: a feature (no business logic), an architecture (no layers/patterns beyond what the single path requires), comprehensive (one path, one test, one entry point).
Procedure
1. Project scaffold
Minimal project structure: init project, configure test runner, create entry point.
- Use the tech stack specified by the user — do not substitute
- Simplest configuration. No linters, formatters, or CI yet — those come with
/pragma:harden - Do NOT install unnecessary dependencies
2. One passing test
A single test proving the pipeline works: import from source, assert something trivially true, run and confirm green. This is a pipeline test, not a feature test.
3. One runnable path
Make the entry point do something minimal and observable:
- Library: export a function callable from the test
- CLI: accept trivial input, produce output
- Server: start, respond to one route, shut down
- UI: render one element
Touch the same boundaries real slices will cross, but with trivial logic. If the capsule has a happy path, create a placeholder.
4. Verify and commit
Run all available checks (typecheck, tests, build, run entry point). Fix before moving on — the skeleton is not done until the full pipeline is green.
Constraints
- No features. The skeleton proves the pipeline, not the product.
- No speculative structure. No directories, modules, or abstractions "for later."
- No dependency shopping. Only what's needed to run. Feature deps come with their slices.
- Tests must actually run and pass. The entry point must actually execute and produce output.
Output
- Pipeline proven: what can now be done (build, test, run)
- Entry point: what it does and how to invoke it
- Test: what it checks and how to run it
- Files created: list
- Next step: typically
/pragma:card→/pragma:slicefor the first real behavior - Lifecycle:
State: foundation,Next: /pragma:card,Loop: /pragma:consult(default unless user explicitly continues directly)
Related Skills
hashintel/petrinaut
documentation
VerifiedTrustedCommunity
Read and write a Petri net (SDCPN) document by Automerge URL. Use when creating, editing, or querying Petri nets — adding or removing places, transitions, arcs, color types, differential equations, and parameters.
34SKILL.mdUpdated May 15, 2026
hashintel/spike
development
VerifiedTrustedCommunity
Time-boxed throwaway investigation to answer one hard question. Use when facing technical uncertainty before a slice — the output is knowledge, not production code. Retires risk by producing a spike verdict with clear recommendations.
34SKILL.mdUpdated May 15, 2026
hashintel/slice
development
VerifiedTrustedCommunity
Implement one tracer-bullet slice following the inside-out methodology. Use when you have a tracer-bullet card ready to build. Implements functional core first, then imperative shell, then end-to-end wiring, then alignment refactor.
34SKILL.mdUpdated May 15, 2026
hashintel/shaping-kickoff
development
VerifiedTrustedCommunity
Turn kickoff transcripts into a builder-facing territory map before carding. Use when kickoff conversations are rich but implementation slices are not yet legible.
34SKILL.mdUpdated May 15, 2026