Workflow for authoring and adapting AI-agent prompts: skills, agents, workflows, rules, etc.
SkillScore77Maintained
Automatic
◆AI Engineer pack
Last verified 8/27/2026
What it changes
Passed 15 of 19 test cases with the skill, and 3 of 19 without it.
Without skill18%
With skill77%
Checks it passed that the model alone missed
Outputs Simulation Notes
Loads pa-adapt.md during adaptation tasks
Assigns gpt-5.6-terra-medium to the discoverer subagent
and 10 more in the full report
13 fixed4 passed either way5 still failing0 broke
Pass-rate lift
+59 pp
Tokens
+49%
Agent turns
0%
Response time
-52%
Model gemini-3.6-flashMeasured 8/27/2026v417 of 22 cases passed with skill · 19 comparable
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 13 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 13
Prompt
Specify the model required for the Blueprint Architect role during step 3 of the prompt authoring workflow.
What was checked, without and with the skill
Checked
Without skill
With skill
Designates gpt-5.6-sol-high for the blueprint architect role
Missed
Passed
case-08 · gemini-3.6-flash · 8/27/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md117 linessingle file
<description_and_purpose>
Orchestrates prompt authoring/adaptation via discover -> extract+intake -> blueprint -> for_each_prompt_loop(draft -> hardening -> edit) -> simulate -> validate. Every phase logs status in coding-agents-prompting-flow-state.md, Prompt Brief carries through all phases, and final output traces to request intent.
</description_and_purpose>
<workflow_phases>
Orchestrator must trust the system and skills; coordinate only sequence, artifacts, state, and approvals.
Load only references needed for the current phase.
Execute phases sequentially, do not skip!
All Rosetta prep steps MUST be FULLY completed
USE SKILL load-project-context, orchestration, hitl
MUST ALWAYS use todo tasks ledger, ASAP. Phases are sequential. Independent tasks can run in parallel.
MUST just-in-time load/execute/update each phase's: instructions, definitions, skills, state file.
Orchestrator and subagents MUST USE SKILL coding-agents-prompt-authoring.
No rush, take your time, MUST FOLLOW WORKFLOW ENTIRELY, no skipping.
This workflow MUST be used with Fable, Opus, GPT-5.5+ class models => IF NOT - DEMAND USER TO SWITCH MODEL.
Workflow state MUST be saved to agents/TEMP/<FEATURE>/coding-agents-prompting-flow-state file.
Discover project-local context, relevant prompt-family artifacts, and required references for this request.