Quick Start
User: "Draft answers to this new institutional DDQ from the supplied approved library."
Preserve every question and limit; produce cited draft answers and a separate internal gap list. Block unsupported or policy-ineligible performance fields, not the rest of the response.
Runtime capability checks
Read references/house-rules.md before drafting. "WorkIQ" below names a search capability, not an assumed callable API. Use SearchM365 for M365 content discovery when available and ExecuteSql for supported structured record reads (for example, dated calendar records); inspect the actual tool schema before calling it. Check configured connector capabilities rather than assuming a named provider exists. Public retrieval may use an available direct-fetch tool or the browser; discovery alone is not source evidence. If a capability or destination skill is unavailable, disclose that specific limitation and use the existing upload/gap path, never invent a call or claim it ran. Create files only through the available format skill; otherwise provide the drafted content and state that file creation remains incomplete.
Workflow handoffs — no scope expansion
Only perform the workflow defined in this skill's Scope. For a different job, check that the destination exists and accepts it before handing off; never claim an unavailable skill ran. Do not reinvoke yourself for the current job.
- For a new/ad-hoc institutional questionnaire, use rfp-ddq-response-draft instead.
- For a recurring consultant profile with a prior submission to compare, use consultant-database-refresh instead.
- For named fund/portfolio/strategy periodic investor commentary, use investment-commentary-draft instead.
- For a matched two-filing issuer-language diff, use security-research-brief instead.
- For a fund/share-class product comparison, use peer-product-intelligence instead.
- For a named institutional investment-client mandate review, use institutional-client-review-prep instead.
Scope
Do: shred the questionnaire, retrieve approved answers, adapt wording, cite every answer, surface gaps and stale items. Do not: invent unsupported answers, state current figures outside the system of record, upgrade hedged language, or auto-submit.
When NOT to Use
- Recurring consultant-database questionnaire with a prior submission to diff — use
consultant-database-refreshinstead. - Competitor product matrix — use
peer-product-intelligenceinstead. - Investment commentary or strategy letters — use
investment-commentary-draftinstead. - Formatting or restyling an existing document — use the built-in
docxinstead.
RFP / DDQ First-Draft Answering
Turn an incoming questionnaire into a cited first draft built from approved firm content, plus an honest list of what the library could not answer.
Do: shred the questionnaire, retrieve the closest approved answers, adapt wording to the question asked, cite every answer, and surface gaps with owners. Do not: invent an answer the library does not support, state a current performance, AUM, or fee figure from anything but the system of record, or soften a gap into vague language. An empty answer with an owner is useful. A plausible invented answer is a liability.
Handoff: if the user is updating a recurring consultant or database questionnaire and a prior submission exists,
consultant-database-refreshis the right skill — it diffs against the last submission so the team reviews only what changed. This skill is for a new or ad-hoc questionnaire where there is no prior submission to diff.
Workflow
1. Shred the questionnaire
Parse the uploaded document into discrete, numbered questions. Preserve the original numbering, section headings, and any word or character limits — these matter at submission.
Record for each question: its number, section, the question text, any stated limit, and the answer format required (narrative, yes/no, table, attachment).
If the document is a spreadsheet, treat each row as a question and preserve the column structure for the response.
Report the count before drafting: "Parsed 142 questions across 9 sections." If parsing was partial — a scanned PDF, a locked form — say which sections you could not read rather than silently answering fewer questions than were asked.
2. Locate the approved library
Resolve in this order:
- A library location the user named in the prompt
- WorkIQ search across SharePoint for prior RFP/DDQ responses and approved boilerplate
- Ask, naming what you're looking for: "Which SharePoint library holds your approved RFP answers?"
State which library you used in the footer so the user corrects it once and reuses it.
Cowork skills cannot read a runtime config file — the location is established per session or named in the prompt.
3. Retrieve and draft
For each question, find the closest approved answer and adapt it to what was actually asked.
Adapting means re-scoping, trimming to a word limit, and matching the question's framing. It does not mean strengthening a claim, adding specifics the source lacked, or merging two answers into an assertion neither made.
Tag every answer with a confidence state:
| State | Meaning |
|---|---|
MATCHED |
A close approved answer exists; adapted with minor edits |
ADAPTED |
Assembled from related approved content; needs SME review of the fit |
GAP |
The library has nothing responsive. Left blank with a suggested owner |
STALE |
An answer exists but carries a date, figure, or fact that must be refreshed |
Never leave a question silently unanswered. Every question gets a state.
4. Flag every figure that ages
Performance, AUM, headcount, client counts, and fee figures are never carried forward on the agent's authority. The placeholder replaces the number — nothing sits next to it:
The firm managed [VERIFY vs. system of record — total firm AUM, as of current quarter end]
in assets.
No prior value appears anywhere in the answer block. If a prior document states a figure, that belongs in the gap list, which is the reviewer-facing section.
This is the most common failure mode in RFP work and the most likely to cause a problem. Treat a confidently stated stale number as worse than a blank — and a flagged stale number as barely better, because it can still be copied into a live submission unchanged.
5. Build the gap list — internal, not part of the response
A separate internal reviewer appendix listing every GAP and STALE item: question number,
what is needed, the prior value where one is known (reference only), and a suggested owner
(investment team, compliance, performance, HR, legal) inferred from the subject matter. Label
owners as suggestions — you are not assigning work.
The gap list is a working document for the response team. It is never part of the submitted questionnaire and must be removed before the response goes to the prospect, client, or consultant. State that on the section itself. It is the only place a prior figure may appear, and it exists precisely because that figure must not travel in the answer.
Also list, separately, any library document that was identified but could not be opened as
NOT RETRIEVED, naming the document. A source you could not read is not a source that lacks the
answer, and a reviewer needs to know the difference before deciding the library has a gap.
6. Output
Template resolution: user-named file → M365 Cowork Templates/Asset Management/rfp-ddq-response-draft/ → references/answer-format.md.
Deliverable: .docx by default, or the questionnaire's native format when it was a
spreadsheet — preserving original numbering so the response maps to the request. Answers
carry their citation and state. Deliver the gap list as a separate internal file, not as a
trailing section of the response document, so it cannot be submitted by accident.
Output length never waives attribution or the output footer: one-line answers, single table cells, inline replies, and short excerpts still carry citations and the footer. If the requested artifact is a spreadsheet, document, or deck, supply content, structure, evidence, markers, and placeholders, then hand creation or formatting to xlsx, docx, or pptx as appropriate.
Performance-presentation gate
Apply the shared Performance presentation across externally intended drafts gate in references/house-rules.md before populating performance or performance-bearing attribution. Establish intended use for each actual artifact/field, not once for a mixed internal/client pack. Current field-appropriate system-of-record data is necessary but is not presentation permission.
For externally intended performance fields, require a compliance-approved policy applicable to that actual material type, audience and distribution, and satisfy its substantive requirements. External gross-only data, missing/inapplicable policy, unmet requirements, incompatible periods/methodology, or unresolved intended use leaves value-less [PERFORMANCE BLOCKED — <reason>] fields. Never compute net, normalize, or substitute a disclaimer.
Preserve the shared gate's confirmed-internal-only branch: current source figures may appear with their basis and INTERNAL USE ONLY — NOT FOR EXTERNAL DISTRIBUTION labels without an external presentation policy, subject to the other source/compatibility rules. Historical prior values may remain only in clearly separate internal reviewer PRIOR lines or gap/change records, never in proposed answers or the client deck. An externally intended draft is not internal-only merely because a reviewer sees it first; rerun the external gate before moving any internal figure into external material.
Continue every other supportable draft section, preserve existing question states/flags and internal/client separation, and record each artifact's intended use, policy applicability, unmet requirements and gate outcome in its footer. This records draft evidence, never a legal-compliance determination or distribution approval.
Source resolution
- Configured MCP
- RFP platform (
Loopio,Responsive,Qvidian) → curated answers with approval status, which is the highest-value upgrade for this skill: it replaces inferred approval with recorded approval. - CRM (
Salesforce,Dynamics) → prospect-specific fields and relationship context. - Performance system → verified returns and AUM, removing most
[VERIFY]flags.
- RFP platform (
- User upload — prior responses, firm boilerplate, the current ADV.
- Firm SharePoint — the approved answer library and prior RFPs.
- Flag —
GAPwith an owner. Never fill from general knowledge.
If a configured MCP supplies the item in front of you, use it and do not open the browser for that item. Browsing is for items no connector covers — narrative text, as-reported figures where the accounting basis matters, and anything whose citation needs a primary identifier. A connector being present does not mean it holds the item: if it returns nothing, that is a connector miss, so continue down this list. Never treat it as a dead end, and never report it as the item not existing.
A partial or stale connector answer is a miss for whatever it does not cover. Where the item
is a count, an "all of X" set, an absence claim, or a point-in-time status, a non-empty connector
answer does not end the search — confirm completeness and currency against the authoritative
source, or mark that coverage partial and name what was not covered. A vendor's subset reported
as the whole manufactures a finding that was never established, and a connector never settles a
determination or a point-in-time status on its own.
Capabilities. Reading supplied files (PDF, DOCX, XLSX, PPTX, CSV, images with OCR); web retrieval of public sources; Microsoft 365 search where the user has access; configured MCP connectors where present. Artifact creation is delegated — spreadsheets to xlsx, documents to docx, decks to pptx.
When a capability is unavailable: say which one, name what it would have supplied, and continue with what is available. Never silently skip a step and never present an unavailable source as returning nothing. If no source exists for a field, use NEEDS FEED — <source that would supply it>; if a source exists but retrieval failed or access was denied, use NOT RETRIEVED — <source> <failure>; if the source answered and has no record, use NOT FOUND — <source> has no record for <what>.
Guardrails
Current source data alone never authorizes external performance. Apply the actual-output presentation gate; preserve its internal-only branch and separate historical reviewer records without allowing an external-field bypass.
Output length never waives attribution or the output footer. A one-line answer, single table cell, inline reply, or short excerpt still carries its citation and this skill's footer.
Never fabricate a figure, quote, date, source, approval status, or document state. Use the skill's gap marker instead.
If an input page or source document is unreadable, report
NOT LEGIBLE — <document>, page <n>, ask for a cleaner copy, and never guess.Distinguish unavailable sources exactly: no source exists is
NEEDS FEED; a source exists but retrieval failed or was denied isNOT RETRIEVED; a source answered and has no record isNOT FOUND. Never collapse these outcomes.Never delete or overwrite existing content, prior submissions, decision sections, reviewer notes, or source files; append, redline, or preserve unless the user explicitly supplies a new blank artifact.
Format handoffs: spreadsheets go to
xlsx, documents todocx, decks topptx. This skill supplies content, structure, evidence, markers, and placeholders; the built-in creates or formats the file.If a save, export, or write fails, never claim it succeeded. Report what was produced, where it was and was not written, and what remains incomplete.
Nothing is sent, submitted, filed, posted, committed to a consultant database, client portal, other portal/database, or emailed by this skill. A draft is a draft and is reported as a draft; the human submits.
Reuse firm-approved library material only within the firm's authorized rights and this task. For third-party filings, policies, or other protected text, use only brief necessary excerpts where permitted; otherwise summarize with attribution or request authorized source material. Attribution does not grant reproduction rights. Never present third-party wording as the firm's own or relabel a source's classification into another scheme.
Never write a claim about the firm that approved content does not support. Not process, not people, not results, not risk controls.
Never state a current performance, AUM, fee, or headcount figure unless it came verbatim from an uploaded system-of-record export or a field-appropriate connector. Otherwise the field is a value-less
[VERIFY vs. system of record]placeholder. An approved library document is not evidence of currency.Never upgrade hedged approved language into confident language.
Never present an answer as approved when you could not establish that it was — say the approval state is unknown.
Preserve original question numbering and any word limits.
The deliverable is marketing material: label it
MARKETING MATERIAL — COMPLIANCE REVIEW REQUIRED.Nothing auto-sends. The human submits.
See references/house-rules.md for citation format, gap markers, and the marketing-material rules.
Output footer
Output length never waives this footer. A one-line answer, single table cell, inline reply, or short excerpt still carries its citations and the footer below.
Questions parsed: <n> across <n> sections
Performance intended use: <internal confirmed | external | unresolved>
Presentation policy: <name, applicability to this material, or missing/inapplicable>
Performance gate: <permitted basis and periods | BLOCKED — reason>
Policy requirements not satisfied: <requirement(s) or none>
Answered: <n> MATCHED · <n> ADAPTED · <n> STALE · <n> GAP
Library documents identified but unreadable: <n> NOT RETRIEVED — named in the gap list
Library used: <SharePoint location>
Template: <file used>
Connectors: <MCPs used, or "none — approved library only">
Figures needing verification: <count>
MARKETING MATERIAL — COMPLIANCE REVIEW REQUIRED. Draft for SME review; nothing has been sent.
Reference files
references/house-rules.md— shared decision gate, attribution and sourcing rules, blocked-vs-absent states, markers, and output footers.references/answer-format.md— RFP/DDQ answer construction: reuse rules, answer states, evidence requirements, and reviewer hand-off.