AC battle-tested-agent
19 production-hardened patterns for AI agents — memory, verification, ambiguity handling, compaction survival, delegation, proof-based handoffs, stale-worker recovery, and self-improvement. Use when hardening an agent for production reliability, when an agent keeps hallucinating or losing context, when handoffs between agents drop details, when delegated work silently fails, or when someone says "my agent is unreliable" or "how do I make this more robust." Works with OpenClaw, Claude Code, Cowork, or any SKILL.md-based agent setup. Includes the Isolated Agent Fabrication Guard plus new delegation hardening patterns for brief quality, completion contracts, acceptance gates, silent-worker recovery, and scoped verifier use.
As a process C 53/100 · Has gaps — weak spots: result and completion, inputs and preconditions, failures and branches
How to improve
- 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 · 0
✓ No critical or high findings
Files scanned: 12. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- note
frontmatter-keyunknown frontmatter key "homepage"
Process rating: all ten parameters 53/100
- 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
- 30Running it twice. 2 mutating operations with no state check
- 40Result and completion. Does not say what the result is
- 50When it triggers. No condition that starts the skill
- 60Tools and files. Uses tools (bash) that frontmatter does not declare
- 100Steps. 36 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 1226 tokens
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)
- +3Output format is not stated: the model decides each time
- +2Single-language instructions
- +5Description quotes 2 example trigger phrases
- +3Description length 730: enough signal without eating the budget
- +4Structure: 10 headings
- +3Step-by-step instructions: 36 items
- +4Has examples (1 code blocks)
- +4Reference files are cited in the instructions (4 of 4)
- +3All 1 scripts are documented
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 96.