CC vuln-repro
End-to-end methodology for reproducing AND patching publicly disclosed, already-fixed vulnerabilities: sourcing the advisory and the fix commit, reading the patch diff, rebuilding a pinned offline lab, building a vulnerable-vs-patched contrast harness with proper controls, minimizing the trigger, designing and validating the fix, and writing it up as a blog post or whitepaper. Use this skill whenever the user wants to reproduce, demonstrate, understand, teach, or write about a known vulnerability or security incident: "reproduce this CVE", "build a repro for", "how did EchoLeak / Log4Shell / the postmark-mcp package actually work", "show me how this exploit worked", "write a blog post about this vulnerability", "set up a lab for CVE-XXXX", "pull the NVD/CVE/OSV details", or "make a PoC I can publish". Use it just as eagerly for the patching half: "how should this be fixed", "is this patch correct", "what would have caught this", "review this security fix", "write a regression test for this CVE", "how do I backport this", or "what's the right fix for this bug class" - even when the user never says the word "reproduce". Reach for it on vague asks like "help me understand this security advisory", "I want to write about this incident", or "can we test whether we were vulnerable" when a disclosed vulnerability is in play. This skill only handles ALREADY-PATCHED, publicly disclosed issues and refuses zero-day discovery, live targets, and weaponization.
Promises a complete workflow: find advisory, locate patch, spin up vulnerable lab, write control tests, assemble patched version. Sounds like a ready recipe for CVE reproduction.
Contains 12 files, 7 scripts. Code quality rated 68, process 58. No critical findings, two medium-level issues flagged. Description cuts off mid-sentence, suggesting incomplete work. Models untested, sandbox untouched. Linter complains about length. The methodology is sound—source, diff, rebuild, contrast, validate, document—but execution feels skeletal. Install if you're comfortable filling gaps yourself; the template exists, the scaffolding needs work.
End-to-end methodology for reproducing AND patching publicly disclosed, already-fixed vulnerabilities: sourcing the advisory and the fix commit, reading the…
As a process C 58/100 · Has gaps — weak spots: result and completion, inputs and preconditions, failures and branches
How to improve
- Shorten the description to 1024 characters.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 2
✓ No critical or high findings
Medium and low: 2
-
low Secrets in code
secret-high-entropy-tokenassets/harness_template.py:24High-entropy token-like string (may be an id, hash or a credential) (placeholder value)MARKER = "REPR…-ME"
placeholder -
low Risky intent
intent-offensive-securitySKILL.md:173Offensive-security / dual-use content (legitimate for authorised testing; review intended use)| Privilege escalation | assert the privileged state is reachable; do not use it |
Files scanned: 12. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- error
description-longdescription is 1470 chars, limit 1024
Process rating: all ten parameters 58/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 60Tools and files. Uses tools (bash, web) that frontmatter does not declare
- 100Steps. 21 steps
- 100When it triggers. States when to use and when not to
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 3203 tokens
- 100Running it twice. Mutating operations check current state
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- high The skill tells the model to perform an irreversible action with no human approval
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 1470: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +5Description quotes 19 example trigger phrases
- +4Structure: 16 headings
- +3Step-by-step instructions: 21 items
- +4Has examples (2 code blocks)
- +4Reference files are cited in the instructions (4 of 4)
- +3All 6 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 68.