BC agent-loop-engineering
Execute an authorized software goal through low-context, bounded-autonomous AI coding loops with persistent state, proactive repair, automatic and functional evidence, layered stage review, independent final acceptance, safe workspace boundaries, and resumable handoffs. Use when a target and acceptance criteria are clear and the user asks to implement, debug, verify, continue autonomously, follow Controller-Developer-QC cycles, resume after context loss, or reduce repeated context and documentation. Typical triggers include keep going, continue where we left off, run it autonomously and fix what breaks, 接着跑 / 继续做 / 自动修 / 断点续跑 / 别再问我每一步, we are stuck on the same failure, 卡在同一个错误上, resume after compaction, 上下文快满了, stop burning tokens on repeated context, run the regression after this repair, and make a flaky test deterministic. Also use for bounded-autopilot, single-writer loop state, failure-signature stop rules, delivery-class-aware evidence, and anti-doc-bloat execution. For vague goals, legacy-state conflicts, requirement discovery, target rebaseline, scope creep triage, or QA acceptance authority, use cms-project-governance first.
Execute an authorized software goal through low-context, bounded-autonomous AI coding loops with persistent state, proactive repair, automatic and functional…
As a process C 51/100 · Has gaps — weak spots: result and completion, when it triggers, inputs and preconditions
How to improve
- Shorten the description to 1024 characters.
- 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: 27. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- error
description-longdescription is 1151 chars, limit 1024
Process rating: all ten parameters 51/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 8 mutating operations with no state check
- 40Result and completion. Does not say what the result is
- 60Tools and files. Uses tools (bash) that frontmatter does not declare
- 100Steps. 53 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 3470 tokens
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 14 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 1151: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- -31 of 3 scripts are never mentioned in SKILL.md
- +1No license
- +2Single-language instructions
- +4Structure: 19 headings
- +3Step-by-step instructions: 53 items
- +4Has examples (5 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 53.