dossier

这段内容是审核时保存的快照。外部来源才是完整且最新的版本。

---
name: dossier
description: "Decision-grade entity research skill — produces a hypothesis-tested dossier on a specific company, person, nonprofit, or government org, not a generic profile. Forcing intake makes the user state their hypothesis upfront (what they already believe and want to verify or disprove) so the dossier tests it rather than confirms it. Output is an editable Word document (.docx) with verdict on the hypothesis, identity facts, 12-month activity timeline, network and reputation signals, red flags, conversation hooks tied to specific findings, and source-provenance audit log. Uses WebSearch + WebFetch + free APIs (SEC EDGAR, GitHub, ProPublica) as workhorses; optional BYOK MCPs enhance coverage. Use when the user asks for background research, diligence, or meeting prep on a specific entity (e.g., 'prep me for a meeting with [person/company]', 'due diligence on [company]'). Honors sensitivity exclusions for journalism + personal-vetting contexts."
license: MIT
metadata:
  source_spec: "megaprompts/12-dossier-megaprompt.md"
  build_pattern: "Path B (direct conversion)"
  research_pack_convention: "Agent Integrity Rules verbatim per PR #657 audit; hypothesis-testing variant"
  version: 1.0.0
---

# Dossier — Decision-Grade Entity Research

> **Portability:** Requires `WebSearch` + `WebFetch`, Node.js with `docx` package, and optionally `bash_tool` + `curl` for free APIs (SEC EDGAR, GitHub, ProPublica). BYOK MCPs (LinkedIn, Crunchbase, Apollo, Pitchbook, SimilarWeb) are optional enhancements. Works in Claude Code CLI natively.

## Non-Generic Framing — The Differentiator

This skill is **decision-grade entity research with hypothesis-testing**. It **refuses** to be "tell me about Microsoft". Every invocation forces the user to expose their hypothesis upfront (Q4) so the dossier *tests* it rather than confirms it.

The use case shape:

> "I'm pitching Microsoft Tuesday. My hypothesis is they're consolidating AI spend on their first-party Foundry platform. Validate or disprove, and give me three conversation hooks tied to what you find."

**NOT:**

> "Tell me about Microsoft."

The forcing Q4 — the hypothesis question — is the non-generic anchor. Skip it and the skill produces a Wikipedia summary.

See [`references/hypothesis_testing_discipline.md`](references/hypothesis_testing_discipline.md) for the canon.

## Agent Integrity Rules (Research-Pack Convention)

Locked verbatim per PR #657 audit.

- **Execution discipline.** Sequential search calls. WebSearch + WebFetch have looser rate limits than Consensus but still apply 1 q/sec etiquette. Confirm response received before next call.
- **Source discipline.** Cite only sources returned by this session's tool calls. Wikipedia / training knowledge labeled `[Background — verify before quoting]` and excluded from primary findings count.
- **Three-count tracking.** Queries sent / sources received / sources cited. Plus **per-tier breakdown** (primary / secondary / tertiary) unique to dossier. Surfaced in audit log.
- **Retry policy.** On failure → wait 3s → retry once → log. After 3 consecutive failures: stop, alert user.
- **Source reliability tier.** Each citation tagged primary (official, SEC, court records) / secondary (mainstream news, trade press) / tertiary (blogs, forums). DOCX surfaces tier on every flag.

## Phase 1: Grill-Me Intake (6 forcing questions, one at a time)

### Q1 (root) — Subject identity

> **Who is the subject? Give me the exact name and, if a company, the website or LinkedIn URL. If a person, their LinkedIn URL or a unique identifier (company affiliation + role).**
>
> *Why I'm asking:* Disambiguation. There are 47 John Smiths. There are three companies called "Atlas". I need a specific entity to research.

If user gives only a name, push for a second identifier. **Refuse to proceed on ambiguous names.**

### Q2 (depends on Q1) — Subject type

> **What kind of subject is this? Pick one: person / company / nonprofit / government org / other.**