AC self-improving-meta
Logs agent-infrastructure issues (prompt drift, rule conflicts, hook failures, context bloat). Default is log-only. Claude/Codex project hooks fire only on explicit meta signals (self-improving-meta, prompt-governance, stackability). The optional OpenClaw agent:bootstrap hook is session-start and not matcher-gated; if enabled it injects a log-only reminder and must not edit files. Does not edit AGENTS.md, SOUL.md, TOOLS.md, MEMORY.md, hooks, or skills unless the user requests a reviewed diff in the current session. Use when prompt-file instructions are misread, a hook fails, a skill does not activate, rules conflict, prompt files waste context, memory is stale, or a skill template is incomplete.
As a process C 63/100 · Has gaps — weak spots: result and completion, inputs and preconditions, running it twice
How to improve
- 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 · 0
✓ No critical or high findings
Files scanned: 15. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 6809 tokens (recommended < 5000); move details to references/
Process rating: all ten parameters 63/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 30Running it twice. 18 mutating operations with no state check
- 60Tools and files. Uses tools (bash) that frontmatter does not declare
- 70Execution cost. Instruction body is 6809 tokens
- 100Steps. 133 steps
- 100When it triggers. States when to use and when not to
- 100Failures and branches. 9 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 22 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
- +1No license
- +2Single-language instructions
- +3Description length 704: enough signal without eating the budget
- +4Structure: 52 headings
- +3Step-by-step instructions: 133 items
- +4Has examples (18 code blocks)
- +4Reference files are cited in the instructions (2 of 3)
- +3All 3 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 81.