JNPRAutomate/fw-skills-share / skills/firewall-config-diff
firewall-config-diff
VerifiedCompare two parsed firewall configurations by semantic intent rather than text. Use when checking drift, HA parity, pre/post-change results, migration fidelity, or round-trip conversion. Parse raw configs first; use a text diff for literal line changes.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 11.2 KB |
| agents/openai.yaml | config | 224 B |
| references/equivalence-rules.md | markdown | 18.1 KB |
| references/example-diff.md | markdown | 15.6 KB |
| references/runtime-intake.md | markdown | 5.4 KB |
references/runtime-intake.md
View on GitHub# Runtime Intake
## Contents
- [When to ask](#when-to-ask)
- [Tool adaptation](#tool-adaptation)
- [Question catalog](#question-catalog)
## When to ask
Use this catalog only after inspecting the request and evidence. Ask an entry
when its `ask_when` condition is true and the answer would materially affect
the result. Skip answered or irrelevant entries. Prioritize safety, scope,
platform or framework basis, evidence quality, then output preference.
## Tool adaptation
- Claude: select at most three neutral entries, project each to only `question`,
`header`, and `options`, then add `multiSelect: false`; do not send `id` or
`ask_when`.
- Codex: select at most three neutral entries and project each to only `id`,
`header`, `question`, and `options`; do not send `ask_when` or `multiSelect`.
- Fallback: ask the same questions in concise plain text with a free-text
`Other` path.
- Never request secrets.
## Question catalog
```json
{
"questions": [
{
"id": "diff_goal",
"ask_when": "Comparison intent is absent.",
"header": "Goal",
"question": "What relationship should the comparison test?",
"options": [
{
"label": "Planned drift (Recommended)",
"description": "Treat A as baseline and B as candidate."
},
{
"label": "HA parity",
"description": "Find unintended peer differences."
},
{
"label": "Migration parity",
"description": "Compare intent across vendors."
}
]
},
{
"id": "diff_direction",
"ask_when": "Input roles are ambiguous.",
"header": "Direction",
"question": "How should ambiguous input roles be resolved?",
"options": [
{
"label": "Establish baseline first (Recommended)",
"description": "Determine the authoritative baseline and comparison direction before classifying changes."
},
{
"label": "Use supplied A-to-B",
"description": "Treat supplied A as baseline and B as new."
},
{
"label": "Compare as peers",
"description": "Treat inputs as unordered and report symmetric differences."
}
]
},
{
"id": "diff_scope",
"ask_when": "Compared components are absent.",
"header": "Scope",
"question": "Which configuration areas should be compared?",
"options": [
{
"label": "All sections (Recommended)",
"description": "Compare every supported component."
},
{
"label": "Policy and NAT",
"description": "Focus on traffic behavior."
},
{
"label": "Named sections",
"description": "Restrict comparison through Other."
}
]
},
{
"id": "diff_identity",
"ask_when": "Rename matching policy affects results.",
"header": "Identity",
"question": "How should renamed elements be matched?",
"options": [
{
"label": "Semantic matching (Recommended)",
"description": "Match resolved meaning before names."
},
{
"label": "Stable names",
"description": "Use names as primary identity."
},
{
"label": "Strict values",
"description": "Report every name or value difference."
}
]
},
{
"id": "diff_ignore",
"ask_when": "Difference allowlist is absent.",
"header": "Exceptions",
"question": "How should an unspecified difference allowlist be handled?",
"options": [
{
"label": "Stop pending allowlist (Recommended)",
"description": "Stop filtering decisions until intentional and generated exceptions are confirmed."
},
{
"label": "Use supplied complete allowlist",
"description": "Exclude every intentional or generated exception in the supplied complete allowlist."
},
{
"label": "Use no exclusions",
"description": "Report all material differences without an allowlist."
}
]
},
{
"id": "diff_output",
"ask_when": "Result detail is absent.",
"header": "Output",
"question": "How should an unspecified result-detail tier be resolved?",
"options": [
{
"label": "Confirm detail first (Recommended)",
"description": "Confirm the result-detail tier before producing the comparison."
},
{
"label": "Use supplied full report",
"description": "Return the complete comparison with all supported difference detail."
},
{
"label": "Use supplied material summary",
"description": "Return only material differences and their risk."
}
]
},
{
"id": "diff_format",
"ask_when": "Result format is absent.",
"header": "Format",
"question": "How should an unspecified result format be resolved?",
"options": [
{
"label": "Confirm format first (Recommended)",
"description": "Confirm the required result format before rendering the comparison."
},
{
"label": "Use supplied human-readable",
"description": "Render the result in the supplied human-readable format."
},
{
"label": "Use supplied machine-readable",
"description": "Render the result in the supplied machine-readable format."
}
]
}
]
}
```
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |