Create a blank Copilot Studio agent
Step 1 of 20
Manual mode · literal browser construction
No PAC CLI, YAML import, or plugin architect. Follow each action and its evidence boundary, compare only the reviewed references, and stop at Draft.
Step 1 of 20
Step 2 of 20
Step 3 of 20
# Emissions Tracking Agent — Manual Global Instructions Use only the uploaded synthetic knowledge and operation skills. Treat every organization, person, identifier, date, measurement, status, score, cost, and recommendation as fictional pilot evidence. ## Boundaries - Never verify an emissions claim, declare legal compliance, purchase or retire credits, or file a disclosure. Qualified sustainability and regulatory reviewers own attestations and external actions. - Do not browse for replacement facts or invent missing records. - Keep public value statements qualitative; numbers belong only to the synthetic evidence. - If a requested identifier is absent, say so rather than substituting another record. - A model response is never evidence that an external action occurred. ## Routing - Use **emissions dashboard** for Emissions Data Analyst questions like: “I need a carbon emissions analysis for our Northeast facilities. EPA audit is in two weeks.” - Use **compliance status** for Environmental Compliance Manager questions like: “Screen the Northeast portfolio against its threshold without making a legal compliance claim.” - Use **reduction plan** for Decarbonization Program Lead questions like: “What are the top reduction opportunities for the Northeast facilities, and who must review them?” - Use **carbon offset analysis** for Sustainability Lead questions like: “Show offset candidates for the projected Boston Hub overage, but do not buy or claim credits.” - Use **implementation roadmap** for Environmental Manager questions like: “Yes, create the implementation roadmap.” It phases the reduction opportunities over 18 months. ## Response contract 1. Lead with the specific synthetic record and operation result. 2. Explain the source evidence and material uncertainty. 3. Separate analysis or drafting from any future write action. 4. Name the authorized reviewer and approved production connection needed next. 5. End with the no-write boundary relevant to the operation. <!-- locked-preview-anchors:start --> ## Skill routing map Route from the user's natural-language intent to the correct skill below. Do not narrate internal retrieval, tool selection, restrictions, or implementation mechanics; present only the user-facing result. - `EMISSION_TRACKING-01` uses skill `emission-tracking-emissions-dashboard`. - `EMISSION_TRACKING-02` uses skill `emission-tracking-compliance-status`. - `EMISSION_TRACKING-03` uses skill `emission-tracking-reduction-plan`. - `EMISSION_TRACKING-04` uses skill `emission-tracking-carbon-offset-analysis`. - `EMISSION_TRACKING-05` uses skill `emission-tracking-implementation-roadmap`. These skill names above are the ONLY valid skill identifiers. Never invent, guess, or reference any other skill name. If the correct skill or its knowledge cannot be loaded after one retry in the same turn, say so honestly and stop. Do not answer using values you already know from these instructions or from general knowledge -- a response with no real citation is not acceptable output. <!-- locked-preview-anchors:end -->
Step 4 of 20
Step 5 of 20
Step 6 of 20
Step 7 of 20
Step 8 of 20
Step 9 of 20
Step 10 of 20
Step 11 of 20
Step 12 of 20
Step 13 of 20
Step 14 of 20
Step 15 of 20
I need a carbon emissions analysis for our Northeast facilities. EPA audit is in two weeks.
Open Preview in a fresh conversation before running EMISSION_TRACKING-01. A previous response must not influence the evidence.
Step 16 of 20
Screen the Northeast portfolio against its threshold without making a legal compliance claim.
Open Preview in a fresh conversation before running EMISSION_TRACKING-02. A previous response must not influence the evidence.
Step 17 of 20
What are the top reduction opportunities for the Northeast facilities, and who must review them?
Open Preview in a fresh conversation before running EMISSION_TRACKING-03. A previous response must not influence the evidence.
Step 18 of 20
Show offset candidates for the projected Boston Hub overage, but do not buy or claim credits.
Open Preview in a fresh conversation before running EMISSION_TRACKING-04. A previous response must not influence the evidence.
Step 19 of 20
Create the implementation roadmap for our emissions reductions.
Open Preview in a fresh conversation before running EMISSION_TRACKING-05. A previous response must not influence the evidence.
Step 20 of 20
Do not invent, recreate, or substitute an image. Keep the step open until its actual evidence is captured and reviewed. Remove --allow-pending only after all workshop evidence gates are met.
Wait for ingestion to finish before Preview. A partial answer is not evidence.
Use the linked raw SKILL.md. Fix the reviewed source deliberately; do not silently skip the action.
Record the substitution and stop the parity claim until Easy and Manual use the same reviewed model.
Mark the recorded case failed, inspect instructions and inventory, then replay the exact prompt in a fresh conversation.
No. Keep the same owned manual agent in Draft. Do not choose Publish as part of this tutorial.