BC self-improving
Self-reflect, self-critique, self-learn, and organize memory with structured logging
Self-reflect, self-critique, self-learn, and organize memory with structured logging
As a process C 60/100 · Has gaps — weak spots: when it triggers, inputs and preconditions, execution cost
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 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.
How to improve
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- 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
-
medium Broad scope
meta-agent-memory-dumpHEARTBEAT.mdAgent memory / workspace files bundled with the skill (2) — likely a workspace dump with personal data or tokensHEARTBEAT.md, memory.md
-
low Risky intent
intent-offensive-securityboundaries.md:13Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)| Access patterns | What systems user has access to | Privilege escalation |
detector
Files scanned: 33. 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") - warning
body-longSKILL.md body ≈ 9088 tokens (recommended < 5000); move details to references/
Process rating: all ten parameters 60/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 11 mutating operations with no state check
- 40Execution cost. Instruction body is 9088 tokens: crowds the task out of the window
- 60Tools and files. Uses tools (bash, web, python) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 100Steps. 294 steps
- 100Failures and branches. 4 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 39 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)
- +3Description length 84: 120–800 characters recommended
- -213 emoji in the instructions: noise for the model
- -32 of 5 scripts are never mentioned in SKILL.md
- +1No license
- +2Single-language instructions
- +4Structure: 147 headings
- +3Step-by-step instructions: 294 items
- +3Output format is stated explicitly
- +4Has examples (63 code blocks)
- +4Reference files are cited in the instructions (3 of 4)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 61.