manager-ai-practices-diligence
Skill69 starsChanged 25 days agoNo licenceView on GitHub
Add an AI-practices module to the diligence file on an external manager — hedge fund, private markets, or long-only — built from the manager's DDQ responses, ODD questionnaire, vendor and service-provider list, policies, and call notes. It maps where AI sits in the manager's investment and operating process: which workflows are model-assisted, what data goes to which outside vendors and under what stated terms, whether outputs that reach an investment decision or an investor document are reviewed by a person before they are relied on, who owns changes to models, prompts, and tools, and what the record of errors and incidents looks like. Each finding is labeled as stated by the manager or evidenced by a document, missing and evasive answers are flagged, and the output is a one-page summary for the diligence file plus the follow-up question set for the manager call. Run on every name in the pipeline and at re-underwriting. For ODD analysts, research analysts, and compliance staff at fund-of-funds, OCIOs, endowments, foundations, pensions, and family offices. Trigger on phrases like "diligence their AI use," "how is this manager using AI," "AI-enabled strategy diligence," "add the AI module to the ODD file," "what are they sending to model vendors," "build the AI questions for the manager call," or when a user attaches a manager's DDQ and asks about AI, machine learning, or model governance. Do NOT use for the full first look at a new manager — use the hedge-fund-diligence skill. Do NOT use as the complete operational review and findings memo — this module is one section that feeds it. Do NOT use for a red-flag read of the whole DDQ — use the ddq-red-flag-scanner skill. Do NOT use to review your own firm's AI workflows — use the ai-workflow-audit skill. Do NOT use to diligence a vendor your own firm is onboarding; this module covers external managers only. Findings describe the manager's practices as stated and evidenced; questions of data licensing or rights are routed to counsel, and nothing here is a legal conclusion or investment advice.
In this repository55 other files it gives its agents.
- ai-workflow-auditSkill
- ambition-upgradeSkill
- call-transcript-followupSkill
- cim-first-passSkill
- claim-sentiment-extractorSkill
- coinvest-triage-memoSkill
- conference-attendee-screenSkill
- conference-prep-plannerSkill
- credit-agreement-covenant-extractorSkill
- ddq-answer-library-curatorSkill
- ddq-red-flag-scannerSkill
- ddq-response-drafterSkill
- deal-pipeline-tracker-updaterSkill
- earnings-call-extractorSkill
- executive-briefingSkill
- expert-call-question-guideSkill
- expert-call-synthesisSkill
- fomc-statement-diffSkill
- fund-doc-consistency-checkerSkill
- fund-marketing-one-pagerSkill
- fund-news-alertSkill
- hedge-fund-diligenceSkill
- hundred-day-plan-drafterSkill
- ic-memo-drafterSkill
- ic-minutes-generatorSkill
- ic-question-anticipatorSkill
- inbox-triageSkill
- intro-email-generatorSkill
- investment-folder-organizerSkill
- investment-podcast-analyzerSkill
- investment-skill-creatorSkill
- key-debates-ledgerSkill
- linkedin-post-generatorSkill
- lp-meeting-prep-briefSkill
- lp-update-generatorSkill
- marketing-compliance-prescreenSkill
- market-map-builderSkill
- market-sizing-builderSkill
- network-referral-path-mapperSkill
- pe-nda-reviewSkill
- performance-attribution-narratorSkill
- principal-background-briefSkill
- professional-profile-briefSkill
- public-company-first-passSkill
- public-co-unit-economicsSkill
- quarterly-letter-synthesizerSkill
- regulatory-exam-binderSkill
- relationship-status-summarizerSkill
- sellside-research-triageSkill
- side-letter-term-extractorSkill
- teach-meSkill
- tear-sheet-private-coSkill
- thesis-red-teamSkill
- waterfall-carry-checkerSkill
- weekly-market-recap-writerSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
No licence file, so all rights are reserved — read it at the source. We count its rules but don't quote them. Read them on GitHub.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

