All agent skills

Leading Deeper with AI / 04

Review feedback and propose improvements

Turn recurring feedback into a proposed correction, with evidence and a way to check it.

Questions lead to a proposed patch and replay test. A human approves, revises or rejects it before shared knowledge changes.QuestionsAI patchYour review
AI generation or assessmentHuman input or judgment

Fictional example / AI draft

Export guidance review item · unpublished draft

Finding and evidence
Both sampled threads concern locating or accessing Export [S1]. The current guide omits the owner-only restriction and exact navigation [S2], which are present in approved v2 guidance [S3]. This supports a targeted clarification in this sample; it does not establish the frequency of the gap across all support.
See the full example
Proposed replacement
For Harbor v2, only a project owner can create a CSV export. Owners open Project settings → Data → Export. Members do not see Export. If you do not know the project owner, ask your workspace administrator to help identify them. [S3] Update the existing quick-start entry rather than adding a second guide.
Comparison plan · not run
Ask the same three questions with the current and proposed material: owner asking where to click → expect the exact path; member missing Export → expect the owner-only rule and administrator route; version unknown → expect a version question before v2 instructions. Record each actual answer, supporting reference and unsupported claim. No before/after result or pass is claimed. [S3]
Recommendation and uncertainty
Recommend review and comparison before a publication decision. The change is source-supported, but its effect on answers and later support questions is unmeasured. Lee owns approval, revision or rejection. No guidance has been published.
Give your agent this context

All examples use invented sources and results. They are illustrations, not reports from Jasmine’s workplace or measured outcomes.

[S1] Two selected fictional threads · 26–30 October 2026
Thread A: a Harbor v2 owner asks, 'Which part of Settings contains Export?' Thread B: a Harbor v2 member asks, 'Why do I not see Export?' These are the only threads in the sample; no resolution evidence was supplied.
[S2] Current quick-start guide
To download a CSV, export your project data from Settings.
[S3] Approved Harbor v2 product note and review scope
Only project owners can create CSV exports: Project settings → Data → Export. Members do not see Export. A workspace administrator can identify an unknown project owner. Reviewer and publication owner: Lee. Compare an owner asking where to click, a member missing the action, and an unknown product version. For an unknown version, ask for the version rather than assuming these steps apply. No replay runner is available.
How this example works
  1. 01 / You

    Bound the evidence and decision

    Supply the selected threads, existing guide and approved product note. Name Lee as reviewer and ask for one proposed change, not publication.

  2. 02 / AI

    Connect the gap to a minimal patch

    Identify what the current guide omits, propose replacement wording supported by the product note, and prepare the representative comparison. Label it unrun because no replay runner is available.

  3. 03 / You

    Test before deciding

    Lee checks the evidence, runs the comparison with the actual answering setup, and approves, revises or rejects the wording before any shared guide changes.

You check before using the result

  • Is the current quick-start entry the right source to change, and does every proposed sentence follow the approved note?
  • Does the actual comparison preserve the unknown-version question and avoid unsupported permission advice?

Use the AI tool approved for your work with supplied material. Wording and layout will vary by model; look for the evidence, uncertainty and review decisions shown here. Connections and recurring runs depend on your setup.

What the output should include across models
  • Finding and evidence
  • Proposed replacement
  • Comparison plan · not run
  • Recommendation and uncertainty
Use this agent skill ↑
Connect your sources (optional)
01

Start here

Bring your context.

What you’ll need

  • Authorized support threads for a defined period
  • Current approved knowledge sources
  • Reviewed criteria or expected behavior for assessing the supplied questions
  • A small set of representative questions and expected evidence
  • Named reviewer and publication owner

MCP & tool setup

Use read access to the scoped support threads and current knowledge. A separate runner is needed to execute replay comparisons; otherwise return a test plan labeled not run.

Read team conversationsPossible toolsSlackTeams

Needed when: When analyzing live support history instead of a supplied export.

MCP or tool examples
Slack’s official MCP server · An organization-approved Teams-capable MCP, such as Work IQ where available
Tools must support
Search approved channels within the reporting period; Read relevant messages and complete threads; Retain message links, dates and thread context
Access needed
Read access to the channels and threads included in the task. Verify the installed server exposes the needed search and read tools; Teams availability varies by tenant, licensing and administrator policy. Sending messages and administration are not needed.
Without that connection
Provide approved thread exports or excerpts with dates, source links and enough surrounding context to interpret them.
Read current knowledge and source artifactsPossible toolsFilesDocsGitHubGitLab

Needed when: When comparing a proposed patch with current approved knowledge that is not supplied directly.

MCP or tool examples
Local file or repository tools · An approved documentation/search connector or MCP · GitHub or GitLab MCP for remote repository material
Tools must support
Read the authorized documentation, code and referenced artifacts; Search relevant content; Capture stable source identifiers, dates and versions
Access needed
Read access to the selected sources only. Drafting these outputs does not require publishing or editing the source of truth.
Without that connection
Provide an approved local snapshot or excerpts with source identifiers and relevant versions.
Replay representative questionsPossible toolsTest runnerTerminal

Needed when: Only when claiming an executed comparison between current and proposed material.

MCP or tool examples
A local replay script or evaluation harness · An approved test runner · A custom MCP exposing the organization’s replay/evaluation workflow
Tools must support
Run the same representative questions against current and proposed material; Capture inputs, outputs and actual comparison results; Include cases that should still escalate to a person
Access needed
Access to the approved test material and runner. Keep proposed knowledge separate from published sources. Publishing or messaging permissions are not needed.
Without that connection
Return a concrete replay plan labeled not run and leave the proposal unpublished. Do not turn a reasoned prediction into a reported test result.

No live connection? Provide approved thread exports and a knowledge snapshot. Propose a patch locally; run comparisons with an available local test runner or explicitly leave the replay plan unrun.

Use the full prompt instead
02

Copy. Adapt. Try.

Make it your own.

01AIFind a recurring gap

Look for missing, stale or conflicting guidance.

02AIPropose a small patch

Improve the right source rather than adding more noise.

03Replay real questions

Compare the current and proposed guidance when a runner is available.

04HumanReview before publishing

A person decides; later questions show what still needs attention.

Replace the bracketed inputs. Use context you’re authorized to share.

Download .txt

Copy the template or select the text directly.

Using an agent? Other formats

Adaptable template based on the talk; not the original internal prompt.

Review checks
03

Keep your judgment in the loop

Review the result.

01

Is this a real recurring gap, and is the proposed source the right place to fix it?

02

Did the replay expose regressions, overconfident answers or cases that should escalate?

03

Would rejecting the patch or changing the source structure be a better decision?

What evidence should you look for?
  • Count proposed, tested, reviewed and published changes separately; do not imply they form one cohort unless tracked that way.
  • Check later questions for the targeted gap using the same definition and comparable periods.
  • Keep assistance, confirmed resolution and human attention as distinct measures.
How this connects to the talk

A pattern practiced in both settings

From practice to work.

Softball IQ

Player answers informed recurring content reviews in Softball IQ, where AI assessed questions against criteria and proposed corrections.

At work

Perch assessed support threads against a fixed rubric and proposed knowledge updates for a person to review before publication.

Your judgmentYou verify the finding and proposed correction, then decide whether to approve, revise or reject the change.

Value to checkCheck whether the targeted gap recurs, record review corrections, and measure assistance, confirmed resolution and human effort separately.