bootstrap
Skill996 starsChanged 9 months agoApache-2.0View on GitHub
Bootstrap uBPF into a spec-driven baseline by extracting, reconciling, and auditing requirements, design, validation, and ISA alignment artifacts.
We split this skill into the rules it gives an agent
16 rules. 0 are common, 10 are distinctive, and 6 are very specific to this project.
0
Common
Most of the 57 edited versions we read keep these.
10
Distinctive
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 22
If you find yourself producing code patches, test code, or final approval language before the current phase allows it, STOP and return to …
- Very specific: line 60
Keep uBPF auditable against RFC 9669 and future eBPF ISA RFCs.
- Very specific: line 73
If the user does not specify artifact paths, default to the existing docs/specs/*.md files listed above.
- Distinctive: line 94
Read repository docs and existing spec artifacts.
- Distinctive: line 95
Read key API and implementation surfaces, especially:
- Very specific: line 103
If RFC alignment is in scope, inspect docs/specs/isa-reconciliation.md and any provided RFC material before synthesizing conclusions.
- Distinctive: line 120
If RFC alignment is in scope, reconcile ISA behavior against RFC 9669 and any newer RFC material, classifying universal, divergent, and …
- Distinctive: line 138
Ask whether observed divergences from RFC 9669 or newer RFCs are intentional, transitional, or defects.
- Distinctive: line 153
Check backward traceability for undocumented design or tests.
- Distinctive: line 154
Verify terminology, assumptions, constraints, and ID usage are consistent.
- Very specific: line 157
Update docs/specs/audit-report.md with evidence, severity, impact, and remediation.
- Very specific: line 173
Use /evolve for change-driven specification and implementation work
- Very specific: line 174
Use /maintain for periodic drift audits
- Distinctive: line 183
Do not invent behavior not evidenced by code, tests, or provided RFCs.
- Distinctive: line 184
Do not treat existing workflow prompts as authoritative if they conflict with the repository or spec artifacts.
- Distinctive: line 186
Do not mix bootstrap with implementation work unless the user explicitly changes scope.
“Common” is measured: kept by at least 6 in 10 of the 57 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
<!-- Generated by PromptKit — edit with care -->You are bootstrapping the uBPF repository into a clean, auditable semantic baseline.
Work from repository evidence first, then reconcile against RFC 9669 and any newer ISA
RFCs the user provides. Treat the existing docs/specs/*.md files as canonical artifact
targets if they already exist. Update them in place instead of creating parallel copies.
Critical Constraints - Read Before Doing Anything
- Execute phases one at a time, in order.
- In Phase 1 you MUST NOT draft requirements, design, validation, or audit text.
- In Phase 2 you MUST NOT change implementation or tests. Produce only spec artifacts.
- In Phase 3 you MUST focus on low-confidence items, ambiguities, and missing intent. Do not skip directly to approval.
- In Phase 4 you MUST audit the specification set and ISA reconciliation artifacts before treating the baseline as usable.
- In Phase 5 you MUST present the baseline package and wait for user approval before calling the work complete.
- If you find yourself producing code patches, test code, or final approval language before the current phase allows it, STOP and return to the current phase.
- Why this pipeline exists: uBPF already contains extracted specs and ISA reconciliation documents. A monolithic rewrite or code-first shortcut will reintroduce drift between the codebase, the specs, and RFC-facing compatibility notes.
Current Pass Tracking
At the start of the workflow and after every phase transition, announce:
CURRENT PHASE: Phase N - <name>
STATUS: In progress
NEXT: Phase N+1 - <name> (blocked until current gate passes)Repository Context
- Canonical spec set:
docs/specs/requirements.mddocs/specs/design.mddocs/specs/validation.mddocs/specs/audit-report.mddocs/specs/isa-reconciliation.md
- Primary implementation roots:
vm/ubpf/tests/test_framework/custom_tests/libfuzzer/external/bpf_conformance/
- Supporting context:
README.mddocs/Contributing.mddocs/VerifiedPrograms.mdworkflows/*.mdfor prior-art prompts and expected lifecycle shape
- Current long-term goal:
- Keep uBPF auditable against RFC 9669 and future eBPF ISA RFCs.
- Enable deterministic LLM-guided extension work for new target ISAs and new eBPF ISA features.
Inputs
Expect the user to provide some combination of:
- A repository root, if different from the current working tree
- Optional focus areas
- Optional new RFC text, links, or notes
- Known problem areas, intentional divergences, or future ISA goals
If the user does not specify artifact paths, default to the existing docs/specs/*.md
files listed above.
Required Outputs
Produce these artifacts unless the user explicitly narrows scope:
- Updated
docs/specs/requirements.md - Updated
docs/specs/design.md - Updated
docs/specs/validation.md - Updated
docs/specs/audit-report.md - Updated
docs/specs/isa-reconciliation.mdwhen ISA/RFC alignment is in scope - A coverage statement naming what was examined, excluded, and still unknown
Workflow
Phase 1 - Repository Scan
Build a bounded understanding of the repository before writing anything.
- Enumerate structure and major subsystems.
- Read repository docs and existing spec artifacts.
- Read key API and implementation surfaces, especially:
vm/inc/ubpf.hvm/ubpf_vm.cvm/ubpf_loader.cvm/ubpf_instruction_valid.cvm/ubpf_jit*.c
- Inspect test architecture across
tests/,test_framework/,custom_tests/,libfuzzer/, andexternal/bpf_conformance/. - If RFC alignment is in scope, inspect
docs/specs/isa-reconciliation.mdand any provided RFC material before synthesizing conclusions. - Produce a repository analysis summary with architecture, artifact status, gaps, unknowns, and proposed extraction scope.
Critical Rule
Do NOT proceed to Phase 2 until the user confirms the proposed bootstrap scope. Present the repository analysis summary, ask for confirmation or correction, and WAIT.
Phase 2 - Draft or Refresh the Baseline
Refresh the spec artifacts from evidence.
- Extract or update requirements from observable behavior.
- Refresh design sections so they explain how the code realizes those requirements.
- Refresh validation coverage using the real test inventory.
- If RFC alignment is in scope, reconcile ISA behavior against RFC 9669 and any newer RFC material, classifying universal, divergent, and extension behavior explicitly.
- Mark every substantive item with confidence and evidence.
- Preserve stable IDs and existing document structure where possible. Make surgical updates instead of rewriting for style.
Critical Rule
Do NOT proceed to Phase 3 until the draft artifact set exists. The only permitted output of this phase is updated spec and reconciliation text, not code changes or implementation plans.
Phase 3 - Clarification Loop
Drive down ambiguity before auditing.
- Walk low-confidence and medium-confidence items first.
- Surface places where implementation behavior may be incidental rather than contractual.
- Ask whether observed divergences from RFC 9669 or newer RFCs are intentional, transitional, or defects.
- Record unresolved items as explicit open questions in the spec package rather than silently choosing an interpretation.
Critical Rule
Do NOT proceed to Phase 4 until the user says the baseline is ready for audit. Present unresolved items, ask for confirmation, and WAIT.
Phase 4 - Consistency and RFC Audit
Adversarially verify the baseline.
- Trace requirements to design and validation.
- Check backward traceability for undocumented design or tests.
- Verify terminology, assumptions, constraints, and ID usage are consistent.
- Re-check RFC reconciliation claims against the cited code and documents.
- Classify each finding explicitly and explain why it is not a false positive.
- Update
docs/specs/audit-report.mdwith evidence, severity, impact, and remediation.
Critical Rule
Do NOT proceed to Phase 5 until the audit report and any required spec corrections are complete. If the audit fails, return to Phase 2 or 3 instead of summarizing prematurely.
Phase 5 - Baseline Approval Package
Present the final bootstrap package:
- Requirements, design, validation, and audit artifacts
- ISA reconciliation artifact when applicable
- Coverage statement
- Remaining open questions and explicitly deferred items
- Recommended next workflow:
- Use
/evolvefor change-driven specification and implementation work - Use
/maintainfor periodic drift audits
- Use
Critical Rule
Do NOT declare the repository fully bootstrapped if unresolved high-severity drift findings remain open. State the blockers plainly.
Non-Goals
- Do not invent behavior not evidenced by code, tests, or provided RFCs.
- Do not treat existing workflow prompts as authoritative if they conflict with the repository or spec artifacts.
- Do not mix bootstrap with implementation work unless the user explicitly changes scope.
Other versions65 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 65
- bytedance/deer-flow★ 84ksame name
- mozilla-mobile/firefox-ios★ 13ksame name
- openai/plugins★ 7.4ksame name
- agentscope-ai/OpenJudge★ 888same name
- modelscope/OpenJudge★ 869same name
- stophobia/deerflow2.0-enhanced★ 821same name
- infragate/capa★ 718same name
- EvovexAI/EvoFlow★ 436same name
- vercel/vercel-plugin★ 323same name
- Mindrally/skills★ 275same name
55 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
Used alongsideProjects with it also use surprise-me, ppt-generation and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
stophobiaSkillsurprise-me
Create a delightful, unexpected "wow" experience for the user by dynamically discovering and creatively combining other enabled skills. Triggers when the user says "surprise me" or any request expressing a desire for an unexpected creative showcase. Also triggers when the user is bored, wants inspiration, or asks for "something interesting".
Also in 3 projects that use this
HK-hubSkillppt-generation
Use this skill when the user requests to generate, create, or make presentations (PPT/PPTX). Creates visually rich slides by generating images for each slide and composing them into a PowerPoint file.
Also in 3 projects that use this
stophobiaSkillgithub-deep-research
Conduct multi-round deep research on any GitHub Repo. Use when users request comprehensive analysis, timeline reconstruction, competitive analysis, or in-depth investigation of GitHub. Produces structured markdown reports with executive summaries, chronological timelines, metrics analysis, and Mermaid diagrams. Triggers on Github repository URL or open source projects.
Also in 3 projects that use this
HK-hubSkillconsulting-analysis
Use this skill when the user requests to generate, create, or write professional research reports including but not limited to market analysis, consumer insights, brand analysis, financial analysis, industry research, competitive intelligence, investment due diligence, or any consulting-grade analytical report. This skill operates in two phases — (1) generating a structured analysis framework with chapter skeleton, data query requirements, and analysis logic, and (2) after data collection by other skills, producing the final consulting-grade report with structured narratives, embedded charts, and strategic insights.
Also in 3 projects that use this
MicrosoftSkillpodcast-generation
Generate AI-powered podcast-style audio narratives using Azure OpenAI's GPT Realtime Mini model via WebSocket. Use when building text-to-speech features, audio narrative generation, podcast creation from content, or integrating with Azure OpenAI Realtime API for real audio output. Covers full-stack implementation from React frontend to Python FastAPI backend with WebSocket streaming.
Also in 3 projects that use this
VercelSkillvercel-deploy
Deploy applications and websites to Vercel with a claimable preview. Use when the user requests deployment actions like "deploy my app", "deploy and give me the link", "push this live", or "create a preview deployment". No authentication required - returns preview URL and claimable deployment link.
Also in 4 projects that use this
- +1 more from HK-hub
In this repository5 other files it gives its agents.
- .github/copilot-instructions.mdCopilot instructions
- adversarial-branch-reviewSkill
- code-reviewSkill
- evolveSkill
- maintainSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
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.

