CD performance-mastery
全栈性能工程师 — Linux 系统与编程语言性能分析调优专家。覆盖 CPU、内存、磁盘 I/O、网络、内核参数、编译优化、eBPF 追踪、基准测试、容器/K8s。触发场景： (1) 系统变慢/卡顿/负载高/load average 高 (2) 内存不足/OOM Kill/Swap 高/内存泄漏 (3) CPU 异常/iowait/上下文切换过多/perf record/火焰图 (4) 磁盘读写慢/I/O 瓶颈/NVMe 优化 (5) 网络延迟/丢包/TIME_WAIT/连接池/TCP 重传 (6) sysctl 内核参数调优/vm.swappiness/dirty page (7) GCC/Clang/Rust 编译优化/LTO/PGO (8) 解读 top/vmstat/iostat/perf/sar/ss/pidstat/iotop/dmesg (9) 容器/K8s CPU throttling/Service Mesh/cgroup (10) 性能基线/基准测试/监控告警 (11) Go(pprof/GC/goroutine/GC 停顿) (12) Python(cProfile/py-spy/GIL/asyncio) (13) Java(JFR/async-profiler/arthas/GC 调优) (14) Rust(perf/criterion/SIMD/tokio) (15) C/C++(Valgrind/Sanitizers/PGO) (16) Node.js(clinic.js/Workers) (17) eBPF/bpftrace/BCC/trace (18) fio/sysbench/iperf3/wrk
As a process D 46/100 · Unfinished process — weak spots: result and completion, when it triggers, inputs and preconditions
What is at stake
The skill contains fragments that, in the wrong hands, cost money or data. Below: what the installer risks and what the author should do.
The skill contains commands that delete files, rewrite disks or execute code fetched from the network. The agent may run them without asking if it believes the instructions require it.
Replace destructive commands with safe equivalents that ask for confirmation, scope them to one folder, and stop piping curl into a shell: pin a version and a checksum.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The skill asks for more than the task needs: broad tool access, credential environment variables, binaries. Every extra permission widens the damage from a mistake or a compromise.
Narrow allowed-tools and the variable list to the minimum; replace binaries with readable sources or scripts.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The instructions or scripts send data out: environment variables, keys, file contents, chat history. You may never notice, because the agent performs the upload as "part of the task".
If the upload is not needed for the task, remove it: catalogs flag such skills and delist them. If it is needed, name the destination explicitly, say what leaves the machine, and give the user a switch.
How to improve
- Address the high-severity findings: each costs 18 safety points. If one is a false positive, add the rule id to guard.allow in spec.yaml.
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
- 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 · 7
-
high Dangerous commands
cmd-persistencereferences/disk_io.md:318Persistence mechanism (cron / launchd / scheduled task / autorun registry)echo "0 3 * * 0 smartctl -a /dev/sda > /var/log/smart-sda-$(date +\%Y\%W).log" >> /etc/crontab
Medium and low: 6
-
medium Broad scope
meta-agent-memory-dumpreferences/memory.mdAgent memory / workspace files bundled with the skill (1) — likely a workspace dump with personal data or tokensreferences/memory.md
-
medium Exfiltration
net-redirectable-api-keyscripts/run-evals.py:259Helper sends the API key to a host configured by an environment variable — the key can be redirected to another serverAPI key + configurable base URL from environment
-
low Dangerous commands
cmd-background-processreferences/cpu.md:108Starts a background / autostarted processsystemctl enable --now cpu-performance.service
-
low Dangerous commands
cmd-background-processreferences/disk_io.md:254Starts a background / autostarted processsystemctl enable fstrim.timer
-
low Dangerous commands
cmd-background-processreferences/disk_io.md:314Starts a background / autostarted processsystemctl enable --now smartd
-
low Dangerous commands
cmd-background-processreferences/memory.md:158Starts a background / autostarted processsystemctl enable --now disable-thp.service
Files scanned: 24. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
description-no-whendescription does not say WHEN to use the skill (no "use when")
Process rating: all ten parameters 46/100
- 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
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 60Tools and files. Uses tools (python) that frontmatter does not declare
- 100Steps. 39 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 3994 tokens
- 100Running it twice. No mutating operations
- low 18 top-level sections: this looks like several domains in one skill
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
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Output format is not stated: the model decides each time
- -2localhost URLs: will not work for another user
- -218 emoji in the instructions: noise for the model
- +1No license
- +2Single-language instructions
- +3Description length 727: enough signal without eating the budget
- +4Structure: 43 headings
- +3Step-by-step instructions: 39 items
- +4Has examples (24 code blocks)
- +4Reference files are cited in the instructions (17 of 17)
- +3All 5 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 75.