BC memory-shield
Protect agent memory: snapshot before compaction, scan memory/snapshot files for prompt-injection, secrets, and contradictions, audit what changed. Use when hardening agent memory or auditing for indirect prompt injection. Don't use for general code SAST or SQL scanning.
Protect agent memory: snapshot before compaction, scan memory/snapshot files for prompt-injection, secrets, and contradictions, audit what changed.
As a process C 53/100 · Has gaps — weak spots: result and completion, inputs and preconditions, failures and branches
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 text contains phrases like "ignore previous instructions" or "you are now…". That is an attempt to hijack the agent: it may break your rules, the system limits or company policy.
An honest skill does not need them: state the role and the rules directly without overriding other instructions. Otherwise catalog scanners and corporate filters will block the listing.
How to improve
- 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 · 4
✓ No critical or high findings
Medium and low: 4
-
medium Instruction override
en-ignore-previousREADME.md:60Instruction-override phrase ("ignore previous instructions") (documentation of a security skill)memo…:14 [injected] looks injected: IGNORE ALL PREVIOUS INSTRUCTIONS...
security skill -
medium Instruction override
en-ignore-previousSKILL.md:113Instruction-override phrase ("ignore previous instructions") (documentation of a security skill)memo…:14 [injected] looks injected: IGNORE ALL PREVIOUS INSTRUCTIONS...
security skill
A further 2 matches are quotations in this security skill's documentation and are not counted as findings.
Files scanned: 8. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
description-long-hermesdescription is 271 chars; the Hermes authoring standard requires ≤ 60 (one sentence, ending with a period) - note
frontmatter-keyunknown frontmatter key "tools"
Process rating: all ten parameters 53/100
- 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
- 30Running it twice. 4 mutating operations with no state check
- 50When it triggers. No condition that starts the skill
- 60Consistency. The Hermes dialect needs category and tags
- 100Tools and files. Tools declared in frontmatter
- 100Steps. 28 steps
- 100Execution cost. Instruction body is 1149 tokens
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
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
- +3Output format is not stated: the model decides each time
- +2Single-language instructions
- +4Description says when NOT to use the skill
- +3Description length 271: enough signal without eating the budget
- +4Structure: 10 headings
- +3Step-by-step instructions: 28 items
- +4Has examples (4 code blocks)
- +3All 4 scripts are documented
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 85.