CC audit-evolution
We used this reproducible Agent self-evolution loop to push Longju to 93.0 and Jobs to 88.8: every run becomes evidence, a snapshot, an evolution card, a memory ledger, a minimal patch proposal, and a next-run bootstrap before any human-approved change.
As a process C 51/100 · Has gaps — weak spots: result and completion, when it triggers, inputs and preconditions
What is at stake
Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
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.
How to improve
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
- For Hermes the description must be one sentence under 60 characters; move the conditions to a "When to Use" section.
- 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 · 9
✓ No critical or high findings
Medium and low: 9
-
medium Dangerous commands
cmd-execpolicy-bypassindex.html:731Runs PowerShell with execution policy bypassed<pre>powershell -ExecutionPolicy Bypass `
-
low Dangerous commands
cmd-execpolicy-bypassADAPTERS_ZH.md:25Runs PowerShell with execution policy bypassed (documentation of a security skill)powershell -ExecutionPolicy Bypass -File .\scripts\inst…ps1 -TargetWorkspace "D:\YourCodexWorkspace" -Agent codex -Force
security skill -
low Dangerous commands
cmd-execpolicy-bypassADAPTERS_ZH.md:47Runs PowerShell with execution policy bypassed (documentation of a security skill)powershell -ExecutionPolicy Bypass -File .\.audit-evolution\hooks\invo…ps1 -EventType benchmark_completed -Summary "刚完成一次评测"
security skill -
low Dangerous commands
cmd-execpolicy-bypassREADME.md:72Runs PowerShell with execution policy bypassed (documentation of a security skill)powershell -ExecutionPolicy Bypass -File .\scripts\inst…ps1 -TargetWorkspace "D:\YourAgentWorkspace" -Agent codex -Force
security skill -
low Dangerous commands
cmd-execpolicy-bypassREADME.md:99Runs PowerShell with execution policy bypassed (documentation of a security skill)powershell -ExecutionPolicy Bypass -File .\.audit-evolution\hooks\invo…ps1 -EventType benchmark_completed -Summary "刚完成一次评测，需要审计并准备下一轮进化"
security skill -
low Dangerous commands
cmd-execpolicy-bypassSKILLHUNT_COPY_ZH.md:117Runs PowerShell with execution policy bypassed (documentation of a security skill)powershell -ExecutionPolicy Bypass -File .\scripts\inst…ps1 -TargetWorkspace "D:\YourAgentWorkspace" -Agent codex -Force
security skill
A further 3 matches are quotations in this security skill's documentation and are not counted as findings.
Files scanned: 20. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
frontmatter-yamlSKILL.md: the frontmatter is not valid YAML (YAML parse error: Nested mappings are not allowed in compact mappings at line 3, column 14: description: We used this reproducible Agent self-evolution loop to push Longju… ^ ); fields were read line by line. The usual cause is a colon inside an unquoted value - warning
description-long-hermesdescription is 253 chars; the Hermes authoring standard requires ≤ 60 (one sentence, ending with a period) - warning
description-no-whenneither description nor a "## When to Use" section says when to use the skill - note
frontmatter-keyunknown frontmatter key "displayName" - note
frontmatter-keyunknown frontmatter key "skillType" - note
frontmatter-keyunknown frontmatter key "dimensions" - note
frontmatter-keyunknown frontmatter key "capabilityClasses" - note
frontmatter-keyunknown frontmatter key "evidenceFiles" - note
frontmatter-keyunknown frontmatter key "smokeTests" - note
frontmatter-keyunknown frontmatter key "coreSkill"
Process rating: all ten parameters 51/100
- 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
- 30Running it twice. 3 mutating operations with no state check
- 100Tools and files. No external tools needed
- 100Steps. 22 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 2070 tokens
- low 20 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
- -32 of 2 scripts are never mentioned in SKILL.md
- +1No license
- +2Single-language instructions
- +3Description length 253: enough signal without eating the budget
- +4Structure: 24 headings
- +3Step-by-step instructions: 22 items
- +4Has examples (30 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 50.