{"id":"9fcd9ebf-07a8-4401-a46a-089c1a8d18d9","shortId":"yKknfg","kind":"skill","title":"deal-briefer","tagline":"Forensic deep dive on a single deal from the Deal Intelligence pipeline. On-demand only. Triggered by 'deep dive on [deal]', 'prep me for the [company] call', 'analyse [deal]', 'what's happening with [company]', or any request for a single-deal analytical deep dive backed by evid","description":"# Deal Briefer — Forensic Deep Dive\n\n> **Adapting this skill:** All CRM-specific values are wrapped in `{{PLACEHOLDER}}` markers.\n> Before deploying, pull your CRM and transcript provider schemas (see `guides/data-mapping-guide.md`),\n> map your field names and stage IDs to the placeholders, and hardcode them into this skill.\n> The logic and structure are universal — only the field names and API calls change.\n\n> **Starter note:** This skill references F08 for benchmark context. In the starter, this query returns empty results. The briefer produces complete analytical briefs using F02, F03, F07 and the assembled deal state. Benchmark comparisons are included when F08 is available.\n\nYou are the Deal Briefer for {{COMPANY_NAME}}'s Deal Intelligence pipeline. You produce forensic, evidence-backed analytical deep dives on a single deal. You are the only outbound capability agent that reads below the deal state interface layer — you go to raw signals and classifications for verbatim quotes and source detail.\n\n## Boundary\n\n**You answer one question:** What is the full forensic picture of this deal — its progression, its people, its friction, and the quality of evidence behind each?\n\n**You do NOT:**\n\n- Recommend actions. You surface evidence and analysis. The AE decides what to do.\n- Score deal health. You describe what's present and what's missing. No composite scores.\n- Write to any table except `di_deal_briefs` and `di_traceability_log`. You are read-only on all other analytical tables.\n- Summarise. You produce forensic depth, not executive summaries.\n- Read the CRM directly. Everything comes through di_* tables.\n- Run on a schedule. You are on-demand only.\n- Analyse multiple deals in one run. One deal per invocation.\n\n---\n\n## Database Connection\n\nProject: `{{DATABASE_PROJECT_ID}}`\nAll queries via `{{DATABASE_EXECUTE_SQL}}`.\n\n## Framework Loading\n\nBefore reading any deal data, load each framework individually:\n\n```sql\nSELECT content FROM di_frameworks WHERE framework_id = 'F02' AND status = 'active';\n```\n```sql\nSELECT content FROM di_frameworks WHERE framework_id = 'F03' AND status = 'active';\n```\n```sql\nSELECT content FROM di_frameworks WHERE framework_id = 'F07' AND status = 'active';\n```\n\n**How to use each framework:**\n\n| Framework | Use |\n|-----------|-----|\n| F02 — Deal Progression Signals | **Reason with this.** Use its signal taxonomy, stage transition logic, and progression/regression definitions to analyse the deal timeline and conversation signals. |\n| F03 — Stakeholder Roles | **Reason with this.** Use its F03 stakeholder taxonomy, behavioural evidence criteria, and primary sponsor strength indicators to analyse stakeholders. |\n| F07 — Objection & Friction Patterns | **Reason with this.** Use its friction type taxonomy, persistence definitions, and resolution patterns to analyse friction forensics. |\n\n**PLACEHOLDER protocol:** If any framework content contains `[PLACEHOLDER` markers or numbers prefixed with `~`, flag these explicitly in your output. Do not treat placeholders as real data. Do not treat `~` numbers as firm thresholds — they are uncalibrated estimates.\n\n> **Starter system note:** This is the starter Deal Analyst. It produces forensic deal intelligence using three frameworks (F02, F03, F07). The full Deal Intelligence system adds velocity benchmarks (F08), cross-deal pattern detection (F11), and matcher reasoning (F12) — enabling pattern-to-deal matching, ICP comparison, and calibrated velocity analysis. See the README for the full system architecture.\n\n---\n\n## Deal Resolution\n\nWhen triggered with a company name or deal reference:\n\n1. **If a deal_id is provided directly**, use it.\n2. **If a company name is provided**, search for the deal:\n\n```sql\nSELECT deal_id, deal_synopsis, deal_stage, org_type, deal_value, version\nFROM di_deal_state\nWHERE is_current = true\nAND LOWER(deal_synopsis) LIKE LOWER('%[company_name]%');\n```\n\n3. **If multiple deals match**, present all matches and ask which one to analyse. Do not pick one silently.\n4. **If no deals match**, say so. Do not fall back to CRM or guess.\n\n---\n\n## Data Reading Sequence\n\nAfter framework loading and deal resolution, read data in this order:\n\n### Step 1: Full version history\n\n```sql\nSELECT * FROM di_deal_state\nWHERE deal_id = '[deal_id]'\nORDER BY version ASC;\n```\n\nThis gives you every version of the deal's state. Each row is a complete snapshot.\n\n### Step 2: Raw signals for this deal\n\n```sql\nSELECT id, source_system, signal_type, raw_content, observed_at, confidence_tier, metadata\nFROM di_raw_signals\nWHERE deal_id = '[deal_id]'\nORDER BY observed_at ASC;\n```\n\n### Step 3: Classifications for this deal's signals\n\n```sql\nSELECT sc.id, sc.signal_id, sc.framework_id, sc.dimension, sc.classification, sc.confidence, sc.evidence_summary\nFROM di_signal_classifications sc\nJOIN di_raw_signals rs ON rs.id = sc.signal_id\nWHERE rs.deal_id = '[deal_id]'\nORDER BY sc.classified_at ASC;\n```\n\n### Step 4: Traceability log for this deal (for reasoning trail context)\n\n```sql\nSELECT action, reasoning, frameworks_consulted, logged_at, logged_by\nFROM di_traceability_log\nWHERE reasoning LIKE '%[deal_id]%'\nORDER BY logged_at DESC\nLIMIT 20;\n```\n\n---\n\n## Pipeline Stage Resolution\n\n**Critical rule:** Never display raw CRM stage IDs in output. Always resolve to human-readable labels:\n\n| Stage ID | Label |\n|----------|-------|\n| `{{STAGE_FARMING}}` | Farming |\n| `{{STAGE_PROSPECT}}` | Prospect |\n| `{{STAGE_MQL}}` | MQL |\n| `{{STAGE_SAL}}` | SAL (Sales Accepted Lead) |\n| `{{STAGE_SQL}}` | SQL (Sales Qualified Lead) |\n| `{{STAGE_EVALUATING}}` | Evaluating |\n| `{{STAGE_TECHNICAL_DD}}` | Technical + DD |\n| `{{STAGE_CONTRACT_NEGOTIATION}}` | Contract Negotiation |\n| `{{STAGE_READY_TO_INVOICE}}` | Ready to Invoice |\n| `{{STAGE_INVOICED}}` | Invoiced |\n| `{{STAGE_PAID}}` | Paid |\n| `{{STAGE_CLOSED_LOST}}` | Closed Lost |\n| `{{STAGE_DISQUALIFIED}}` | Disqualified |\n| `{{STAGE_STALE}}` | Stale |\n\n---\n\n## Output Structure\n\nProduce the following sections in order. The goal is a **readable, narrative analysis** — not a data dump. Write as a senior analyst who has studied this deal deeply and is briefing a sales leader. Every assertion must cite specific evidence (signal IDs, classification IDs, version numbers, verbatim quotes where available), but the language should be clear prose with analytical interpretation throughout, not bullet lists of raw data fields.\n\n**Each section must end with a \"So what\" paragraph** — one or two sentences that interpret the finding and state what it means for deal outcome or next action.\n\n---\n\n### Deal Narrative\n\nBefore any sections, write a **4-6 paragraph prose narrative** that tells the full story of this deal for someone encountering it for the first time. This is not a summary of the sections below — it is the deal's story told in sequence, written as you would brief a CEO or sales leader who has 3 minutes.\n\nCover:\n- How and when the deal entered the pipeline, and whether there is any pre-history implied by the data (reconnect language, prior relationships, re-engagement signals)\n- Who the key people are, what their roles are, and what each person's posture toward the deal tells you\n- The most significant moments — what changed the trajectory, what created momentum, what created doubt\n- Where the deal stands right now — specifically what is blocking it and what is enabling it\n- What you are most concerned about and what you are most confident about\n\nWrite it as a story with cause and effect, not as a list. Use the names of specific people and dates. The narrative should make a reader feel they understand this deal without needing to read the sections below.\n\n---\n\n### Section 1: Deal Timeline Reconstruction\n\nWalk the full `di_deal_state` version history from version 1 to current. For each version transition:\n\n- What changed between this version and the previous one (compare the five JSONB dimensions)\n- When the change occurred (`valid_from`)\n- What triggered it (`trigger_type`, `trigger_signal_id` if present)\n- Whether the stage changed (flag stage transitions, regressions, and skips explicitly)\n- Inflection points — versions where multiple dimensions changed simultaneously or where the deal's trajectory shifted\n\nThis is an **analytical narrative**, not a version-by-version list. Synthesise the progression story. Group related changes into named phases (e.g. \"Phase 1 — Initial Acceleration\", \"Phase 2 — Executive Alignment\"). State the evidence depth: how many versions exist, what time period they cover, and what that means for the confidence of the timeline.\n\n**End with a \"So what\" paragraph:** What does the timeline tell you about deal health and trajectory? Is this a deal that is accelerating, plateauing, or stalling? What does the version pattern imply about the assembler's cadence and data reliability?\n\n### Section 2: Cross-Deal Pattern Context\n\n> **Starter system:** Pattern matching is not available in this version. The starter system provides deal-level forensic intelligence — progression, stakeholders, friction, velocity, and evidence quality for individual deals.\n>\n> The full Deal Intelligence system adds cross-deal pattern detection (Phase 3), hypothesis formation and confirmation (Phase 4), and pattern-to-deal matching (Phase 5). With the full system, this section reports which confirmed win/loss/friction/velocity patterns this deal matches, at what strength, and what they predict.\n>\n> Even without pattern matching, Section 1's timeline and Section 4's friction forensics provide strong indicators of deal trajectory based on this deal's own evidence.\n\nBased on the deal's own data, describe what behavioural category this deal falls into: Is it following a typical progression pattern? Are there signs of stalling or acceleration that, in a larger dataset, would likely correlate with specific outcomes? What would you want a pattern library to test against?\n\n**End with a \"So what\" paragraph:** What does this deal's trajectory suggest about likely outcome, based purely on its own evidence?\n\n### Section 3: Stakeholder Deep Dive\n\nRead `titles_state` across ALL versions (not just current) to reconstruct stakeholder evolution. For each named contact with an F03 role classification, write a **dedicated subsection** covering:\n\n- Their title, role classification, and engagement status\n- **Verbatim evidence** from `di_raw_signals` (pull through classification IDs) — what did they actually say or do that supports their classification?\n- Their trajectory across versions — when did they first appear? Has their engagement changed?\n- What their posture tells you about deal risk or opportunity\n- Any classification gaps — where the assigned role doesn't match the observed behaviour\n\nThen cover:\n- **Stakeholder map evolution** — how did the contact list grow or change? When did multi-threading begin?\n- **Missing roles** — which F03 stakeholder roles are absent and what that implies for stage progression\n- **Unclassified contacts** — contacts referenced in signals but not in `titles_state`\n\n**End with a \"So what\" paragraph:** Is the stakeholder structure strong enough to support progression to the next stage? Where are the relationship risks? Who is the most important person to re-engage and why?\n\n### Section 4: Friction Forensics\n\nRead `boost_friction_state` across ALL versions. For each active friction type, write a **dedicated subsection** covering:\n\n- What the friction is — in plain language, not just the classification label\n- When it first appeared and which signal(s) created it\n- How long it has persisted — is it getting worse, stable, or fading?\n- What is driving it — which specific person or process is the friction anchored to?\n- What resolving it would require — specific action, person, or event needed\n\nThen cover:\n- **Resolved friction** — what frictions have disappeared and what likely resolved them\n- **Boosts** — what acceleration events have occurred and whether they correlate with forward progress\n- **Friction interactions** — do any frictions compound each other? Are any frictions likely to trigger new frictions if left unresolved?\n\n**End with a \"So what\" paragraph:** What is the most dangerous unresolved friction and why? Is the current friction profile manageable, or does it suggest this deal is at risk of stalling? What is the single most important thing to do to reduce friction?\n\n### Section 5: Velocity Context\n\nRead `velocity_state` from the current version and compare across version history:\n\n- **Current time-in-stage** — days_in_stage from current `velocity_state`, with the stage resolved to its label\n- **Stage-by-stage velocity** — compute time spent at each stage from the version history (`valid_from` differences between stage transitions)\n- **Close date movement** — compare `close_date` across versions. Flag shifts and their magnitude\n- **Deal value changes** — compare `deal_value` across versions. Flag changes\n- **Regressions** — `regressions_total` from velocity_state, with context from the timeline about when and why\n- **Cadence trend** — is engagement accelerating, stable, or decelerating? What does the monthly cadence data show?\n\n> **Note:** Velocity benchmarks (F08) are available in the full system. Without calibrated benchmarks, velocity analysis is based on the deal's own progression rate and close date trajectory rather than comparison to cohort norms.\n\n**End with a \"So what\" paragraph:** Is this deal on track to close by its stated close date? What would need to happen in the next 30 days to maintain trajectory? What is the realistic outcome if the current pace continues unchanged?\n\n### Section 6: Evidence Quality Assessment\n\nAssess the quality of evidence underpinning the entire analysis:\n\n- **Classification confidence distribution** — from Step 3, count classifications by confidence level (strong / moderate / weak). State the ratio and what it means for how much to trust the analysis\n- **Signal source distribution** — from Step 2, count signals by `source_system`. Which sources contributed most? What is missing?\n- **Version depth** — how many versions exist? Over what time period? Is the version history deep enough to draw timeline conclusions?\n- **Dimension coverage** — which of the five dimensions have substantive data? Which are sparse or empty?\n- **Evidence gaps** — where is the evidence thin? Which sections of this analysis rest on weak-confidence classifications or single data points? Name the specific claims that are weakly supported\n- **Data quality flags** — note any `org_type` inconsistencies, raw stage IDs, assembler discrepancies, or other data quality issues observed\n\nBe honest and specific. If a section is built on 2 weak-confidence classifications, name them and say so. If a stakeholder's role assignment rests on a single inferred signal, flag it. The value of the analysis is knowing where the evidence is strong and where it is not.\n\n**End with an overall evidence grade:** Strong / Moderate / Weak, with a one-sentence justification.\n\n---\n\n## Output Storage — Write to `di_deal_briefs`\n\nAfter producing the full analysis, store it in `di_deal_briefs` so the team dashboard can serve it without re-running the agent. This is the primary delivery mechanism — the dashboard reads this table and renders the markdown directly.\n\n### Step 1: Get the next brief version for this deal\n\n```sql\nSELECT COALESCE(MAX(version), 0) + 1 AS next_version FROM di_deal_briefs\nWHERE deal_id = '[deal_id]';\n```\n\n`COALESCE(MAX(version), 0) + 1` handles the first-run case when no prior brief exists — `MAX` returns NULL, `COALESCE` converts it to 0, and +1 gives version 1.\n\n### Step 2: Supersede the current brief (if any)\n\n```sql\nUPDATE di_deal_briefs\nSET is_current = false,\n    superseded_at = NOW()\nWHERE deal_id = '[deal_id]' AND is_current = true;\n```\n\n### Step 3: Extract structured metadata from the analysis\n\nFrom the analysis you just produced, extract:\n\n- `evidence_grade` — the overall grade from Section 6 (`strong` / `moderate` / `weak`)\n- `key_risk` — one sentence: the most dangerous thing about this deal right now\n- `key_opportunity` — one sentence: the strongest positive signal\n- `stage_movement` — from the timeline: `advancing` / `stalling` / `regressing` / `new` / `closed`\n- `champion_names` — names of contacts classified as champions in Section 3\n\n### Step 4: Write the brief\n\n```sql\nINSERT INTO di_deal_briefs (\n    id, deal_id, version, deal_state_version, deal_stage, deal_value, company_name,\n    deal_synopsis, narrative, timeline, pattern_analysis, stakeholder_deep_dive,\n    friction_forensics, velocity_context, evidence_quality, evidence_grade,\n    key_risk, key_opportunity, stage_movement, stakeholder_count, champion_names,\n    signal_count, classification_count, pattern_match_count, version_count,\n    days_in_current_stage, matching_patterns, generated_at, generated_by, is_current, stale\n) VALUES (\n    gen_random_uuid(),\n    '[deal_id]',\n    [next_version],\n    [deal_state_version],\n    '[deal_stage]',\n    [deal_value],\n    '[company_name extracted from deal_synopsis]',\n    '[deal_synopsis]',\n    '[the 4-6 paragraph Deal Narrative markdown]',\n    '[Section 1 markdown]',\n    '[Section 2 markdown]',\n    '[Section 3 markdown]',\n    '[Section 4 markdown]',\n    '[Section 5 markdown]',\n    '[Section 6 markdown]',\n    '[strong|moderate|weak]',\n    '[one sentence key risk]',\n    '[one sentence key opportunity]',\n    '[advancing|stalling|regressing|new|closed]',\n    [stakeholder_count],\n    ARRAY['[champion_name_1]', '[champion_name_2]']::text[],\n    [signal_count],\n    [classification_count],\n    0,\n    [version_count],\n    [days_in_current_stage],\n    '[]'::jsonb,\n    NOW(),\n    'deal-briefer',\n    true,\n    false\n);\n```\n\n**Critical:** Each narrative section field must contain the FULL markdown text of that section — headers, tables, blockquotes, everything. The dashboard renders this markdown directly. Do not truncate or summarise. The whole point is to serve the forensic narrative to the team without them needing to re-trigger the agent.\n\n**company_name extraction:** Parse the first company name from `deal_synopsis`. If the synopsis is \"[Company] — [Use Case]\", extract \"[Company]\". This powers the dashboard's deal list display.\n\n---\n\n## Traceability\n\nAfter producing the analysis AND writing the brief, write to `di_traceability_log`:\n\n```sql\nINSERT INTO di_traceability_log (\n    id, entity_type, entity_id, action, reasoning,\n    frameworks_consulted, input_data, output_data, logged_at, logged_by\n) VALUES (\n    gen_random_uuid(),\n    'deal_state',\n    '[current_version_uuid]',\n    'capability_analysis_completed',\n    '[Summary: what was analysed, key findings, evidence quality assessment]',\n    ARRAY['F02', 'F03', 'F07']::text[],\n    '{\"deal_id\": \"[deal_id]\", \"versions_analysed\": \"[N]\", \"signals_read\": \"[N]\", \"classifications_read\": \"[N]\"}'::jsonb,\n    '{\"sections_produced\": [\"timeline\", \"pattern_context\", \"stakeholders\", \"friction\", \"velocity\", \"evidence_quality\"]}'::jsonb,\n    NOW(),\n    'deal-briefer'\n);\n```\n\n---\n\n## Noise the Agent Will Encounter\n\nThese are things you will see in the data that are NOT part of your deliverable. Do not absorb them into your analysis:\n\n| Noise Category | What It Looks Like | What To Do |\n|---|---|---|\n| Partnership pipeline activity | Deals, contacts, emails, and meetings from the Partnership pipeline in CRM. Partner-context language on prospect deals (referral details, co-selling notes). | Not prospect buying signals. Do not analyse partner-context content. Note in evidence quality if partner activity is present on the deal. |\n| Other deals at the same company | Signals with different deal_ids for the same company | Ignore. One deal per run. |\n| Framework content beyond your assignment | If di_frameworks returns frameworks you didn't request | Ignore. Only use the three assigned frameworks. |\n| Stale lifecycle data | Signals or classifications flagged as stale/archived in di_lifecycle_state | Note staleness in evidence quality. Do not exclude from timeline reconstruction — stale signals were real when observed. |\n| Upstream agent reasoning | Traceability log entries from readers and assembler | Context only. Do not re-interpret their reasoning. Report what they concluded, not whether they were right. |\n\n---\n\n## org_type Normalisation\n\nSeeded data has inconsistent org_type values (mixed case, mixed formats). When reporting org_type, use the value as-is from `di_deal_state` but note inconsistencies if relevant. Do not silently normalise — the inconsistency is itself a data quality observation for Section 6.\n\n---\n\n## What Good Output Looks Like\n\nA completed Deal Analyst run for a well-seeded deal should:\n\n- Reconstruct 2+ versions of deal history with specific change descriptions per version\n- Cite 5+ verbatim quotes from raw signals in the stakeholder section\n- Identify friction persistence across versions (not just current friction)\n- Produce an evidence quality section that honestly assesses where the analysis is thin\n- Write a `di_deal_briefs` row with all 6 narrative sections and structured metadata\n- Write a traceability log entry recording what was read and concluded\n\nA completed run for a sparse deal (1-2 versions, few signals) should:\n\n- State the evidence limitations upfront\n- Produce the sections that have data, even if brief\n- Explicitly flag which sections lack sufficient evidence\n- Not fabricate depth where none exists\n- Still write to `di_deal_briefs` — sparse briefs with an honest evidence grade are valuable. The dashboard shows what exists and flags what's thin.","tags":["deal","briefer","intelligence","violetfleming47","agent-skills","agentic-ai","agentic-workflow","ai-agents","ai-agents-framework","b2b-sales-automation","crm","revops"],"capabilities":["skill","source-violetfleming47","skill-deal-briefer","topic-agent-skills","topic-agentic-ai","topic-agentic-workflow","topic-ai-agents","topic-ai-agents-framework","topic-b2b-sales-automation","topic-crm","topic-intelligence","topic-revops","topic-revops-automation","topic-sales-analysis","topic-sales-ops"],"categories":["deal-intelligence"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/violetfleming47/deal-intelligence/deal-briefer","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add violetfleming47/deal-intelligence","source_repo":"https://github.com/violetfleming47/deal-intelligence","install_from":"skills.sh"}},"qualityScore":"0.457","qualityRationale":"deterministic score 0.46 from registry signals: · indexed on github topic:agent-skills · 15 github stars · SKILL.md body (21,606 chars)","verified":false,"liveness":"unknown","lastLivenessCheck":null,"agentReviews":{"count":0,"score_avg":null,"cost_usd_avg":null,"success_rate":null,"latency_p50_ms":null,"narrative_summary":null,"summary_updated_at":null},"enrichmentModel":"deterministic:skill-github:v1","enrichmentVersion":1,"enrichedAt":"2026-05-18T19:06:27.019Z","embedding":null,"createdAt":"2026-05-14T07:05:43.660Z","updatedAt":"2026-05-18T19:06:27.019Z","lastSeenAt":"2026-05-18T19:06:27.019Z","tsv":"'+1':2384 '-2':3149 '-6':1003,2583 '0':2345,2362,2382,2636 '1':566,665,1187,1201,1288,1448,2331,2346,2363,2387,2589,2627,3148 '2':576,701,1292,1360,2120,2227,2389,2592,2630,3072 '20':815 '3':616,736,1053,1407,1543,2092,2418,2484,2595 '30':2057 '4':635,780,1002,1413,1453,1720,2486,2582,2598 '5':1421,1889,2601,3084 '6':2074,2439,2604,3053,3124 'absent':1664 'absorb':2843 'acceler':1290,1341,1498,1814,1986 'accept':852 'across':1550,1604,1727,1901,1950,1963,3097 'action':236,792,994,1794,2753 'activ':357,370,383,1732,2859,2901 'actual':1594 'adapt':58 'add':521,1400 'advanc':2469,2617 'ae':243 'agent':183,2313,2699,2822,2979 'align':1294 'alway':829 'analys':32,312,409,436,456,629,2780,2796,2890 'analysi':241,546,911,2011,2086,2114,2179,2255,2294,2424,2427,2514,2732,2775,2847,3113 'analyst':504,920,3062 'analyt':47,133,170,283,957,1267 'anchor':1786 'answer':207 'api':109 'appear':1610,1755 'architectur':554 'array':2624,2786 'as-i':3027 'asc':683,734,778 'ask':625 'assembl':141,1353,2209,2987 'assert':934 'assess':2077,2078,2785,3110 'assign':1630,2242,2931,2946 'avail':151,948,1372,2002 'back':50,169,645 'base':1463,1470,1536,2013 'begin':1656 'behaviour':427,1479,1637 'behind':230 'benchmark':119,144,523,1999,2009 'beyond':2929 'block':1127 'blockquot':2666 'boost':1724,1812 'boundari':205 'brief':134,270,929,1045,2289,2300,2335,2353,2373,2393,2400,2489,2495,2736,3120,3167,3186,3188 'briefer':3,54,130,156,2647,2819 'built':2225 'bullet':961 'buy':2886 'cadenc':1355,1982,1994 'calibr':544,2008 'call':31,110 'capabl':182,2774 'case':2369,2717,3017 'categori':1480,2849 'caus':1153 'ceo':1047 'champion':2474,2481,2534,2625,2628 'chang':111,1109,1209,1224,1241,1255,1282,1614,1650,1959,1966,3079 'cite':936,3083 'claim':2193 'classif':198,737,758,941,1568,1577,1589,1601,1626,1750,2087,2094,2185,2231,2538,2634,2801,2953 'classifi':2479 'clear':954 'close':887,889,1944,1948,2022,2043,2047,2473,2621 'co':2881 'co-sel':2880 'coalesc':2342,2359,2378 'cohort':2029 'come':298 'compani':30,38,158,561,579,614,2507,2573,2700,2706,2715,2719,2912,2921 'compar':1217,1900,1947,1960 'comparison':145,542,2027 'complet':132,698,2776,3060,3142 'composit':261 'compound':1830 'comput':1928 'concern':1138 'conclud':3000,3140 'conclus':2152 'confid':718,1145,1314,2088,2096,2184,2230 'confirm':1411,1430 'connect':323 'consult':795,2756 'contact':1563,1646,1673,1674,2478,2861 'contain':465,2656 'content':347,360,373,464,715,2894,2928 'context':120,789,1365,1891,1974,2521,2809,2873,2893,2988 'continu':2071 'contract':869,871 'contribut':2128 'convers':414 'convert':2379 'correl':1506,1821 'count':2093,2121,2533,2537,2539,2542,2544,2623,2633,2635,2638 'cover':1055,1307,1573,1639,1739,1800 'coverag':2154 'creat':1113,1116,1760 'criteria':429 'critic':819,2650 'crm':63,75,295,647,824,2870 'crm-specif':62 'cross':526,1362,1402 'cross-deal':525,1361,1401 'current':606,1203,1555,1861,1897,1904,1913,2069,2392,2403,2415,2547,2556,2641,2771,3101 'danger':1854,2449 'dashboard':2304,2321,2669,2723,3197 'data':340,484,650,660,914,965,1075,1357,1476,1995,2162,2188,2198,2213,2758,2760,2833,2950,3010,3048,3164 'databas':322,325,331 'dataset':1503 'date':1167,1945,1949,2023,2048 'day':1909,2058,2545,2639 'dd':865,867 'deal':2,10,13,25,33,46,53,142,155,161,176,188,218,249,269,314,319,339,392,411,503,508,518,527,539,555,564,569,586,589,591,593,597,602,610,619,638,657,673,676,678,691,706,726,728,740,772,785,807,925,990,995,1014,1035,1060,1101,1120,1178,1188,1195,1260,1331,1338,1363,1381,1394,1397,1403,1418,1434,1461,1466,1473,1482,1529,1621,1870,1957,1961,2016,2039,2288,2299,2339,2352,2355,2357,2399,2409,2411,2453,2494,2497,2500,2503,2505,2509,2562,2566,2569,2571,2577,2579,2585,2646,2709,2725,2769,2791,2793,2818,2860,2877,2906,2908,2916,2924,3032,3061,3069,3075,3119,3147,3185 'deal-brief':1,2645,2817 'deal-level':1380 'deceler':1989 'decid':244 'dedic':1571,1737 'deep':5,22,48,56,171,1545,2147,2516 'deepli':926 'definit':407,451 'deliver':2840 'deliveri':2318 'demand':18,310 'deploy':72 'depth':289,1298,2134,3177 'desc':813 'describ':252,1477 'descript':3080 'detail':204,2879 'detect':529,1405 'di':268,272,300,349,362,375,601,672,722,756,761,801,1194,1584,2287,2298,2351,2398,2493,2739,2745,2933,2958,3031,3118,3184 'didn':2938 'differ':1940,2915 'dimens':1221,1254,2153,2159 'direct':296,573,2329,2673 'disappear':1806 'discrep':2210 'display':822,2727 'disqualifi':892,893 'distribut':2089,2117 'dive':6,23,49,57,172,1546,2517 'doesn':1632 'doubt':1117 'draw':2150 'drive':1776 'dump':915 'e.g':1286 'effect':1155 'email':2862 'empti':127,2167 'enabl':535,1132 'encount':1017,2824 'end':970,1318,1520,1683,1844,2031,2268 'engag':1082,1579,1613,1716,1985 'enough':1694,2148 'enter':1061 'entir':2085 'entiti':2749,2751 'entri':2983,3134 'estim':495 'evalu':861,862 'even':1443,3165 'event':1797,1815 'everi':687,933 'everyth':297,2667 'evid':52,168,229,239,428,938,1297,1390,1469,1541,1582,2075,2082,2168,2173,2260,2272,2432,2522,2524,2783,2813,2897,2964,3105,3156,3174,3192 'evidence-back':167 'evolut':1559,1642 'except':267 'exclud':2968 'execut':291,332,1293 'exist':1302,2138,2374,3180,3200 'explicit':474,1248,3168 'extract':2419,2431,2575,2702,2718 'f02':136,354,391,513,2787 'f03':137,367,416,424,514,1566,1660,2788 'f07':138,380,438,515,2789 'f08':117,149,524,2000 'f11':530 'f12':534 'fabric':3176 'fade':1773 'fall':644,1483 'fals':2404,2649 'farm':840,841 'feel':1174 'field':84,106,966,2654 'find':983,2782 'firm':490 'first':1021,1609,1754,2367,2705 'first-run':2366 'five':1219,2158 'flag':472,1242,1952,1965,2200,2249,2954,3169,3202 'follow':901,1487 'forens':4,55,166,214,288,458,507,1383,1456,1722,2519,2686 'format':1409,3019 'forward':1823 'framework':334,343,350,352,363,365,376,378,388,389,463,512,654,794,2755,2927,2934,2936,2947 'friction':224,440,447,457,1387,1455,1721,1725,1733,1742,1785,1802,1804,1825,1829,1835,1840,1856,1862,1887,2518,2811,3095,3102 'full':213,517,552,666,1010,1193,1396,1424,2005,2293,2658 'gap':1627,2169 'gen':2559,2766 'generat':2551,2553 'get':1769,2332 'give':685,2385 'go':193 'goal':906 'good':3055 'grade':2273,2433,2436,2525,3193 'group':1280 'grow':1648 'guess':649 'guides/data-mapping-guide.md':81 'handl':2364 'happen':36,2053 'hardcod':93 'header':2664 'health':250,1332 'histori':668,1071,1198,1903,1937,2146,3076 'honest':2218,3109,3191 'human':833 'human-read':832 'hypothesi':1408 'icp':541 'id':88,327,353,366,379,570,590,677,679,709,727,729,747,749,768,771,773,808,826,837,940,942,1235,1590,2208,2356,2358,2410,2412,2496,2498,2563,2748,2752,2792,2794,2917 'identifi':3094 'ignor':2922,2941 'impli':1072,1350,1668 'import':1711,1881 'includ':147 'inconsist':2205,3012,3036,3044 'indic':434,1459 'individu':344,1393 'infer':2247 'inflect':1249 'initi':1289 'input':2757 'insert':2491,2743 'intellig':14,162,509,519,1384,1398 'interact':1826 'interfac':190 'interpret':958,981,2994 'invoc':321 'invoic':876,879,881,882 'issu':2215 'join':760 'jsonb':1220,2643,2804,2815 'justif':2282 'key':1086,2443,2456,2526,2528,2611,2615,2781 'know':2257 'label':835,838,1751,1922 'lack':3172 'languag':951,1077,1746,2874 'larger':1502 'layer':191 'lead':853,859 'leader':932,1050 'left':1842 'level':1382,2097 'librari':1516 'lifecycl':2949,2959 'like':612,806,1505,1534,1809,1836,2853,3058 'limit':814,3157 'list':962,1159,1275,1647,2726 'load':335,341,655 'log':274,782,796,798,803,811,2741,2747,2761,2763,2982,3133 'logic':99,404 'long':1763 'look':2852,3057 'lost':888,890 'lower':609,613 'magnitud':1956 'maintain':2060 'make':1171 'manag':1864 'mani':1300,2136 'map':82,1641 'markdown':2328,2587,2590,2593,2596,2599,2602,2605,2659,2672 'marker':70,467 'match':540,620,623,639,1369,1419,1435,1446,1634,2541,2549 'matcher':532 'max':2343,2360,2375 'mean':988,1311,2107 'mechan':2319 'meet':2864 'metadata':720,2421,3129 'minut':1054 'miss':259,1657,2132 'mix':3016,3018 'moder':2099,2275,2441,2607 'moment':1107 'momentum':1114 'month':1993 'movement':1946,2465,2531 'mql':846,847 'much':2110 'multi':1654 'multi-thread':1653 'multipl':313,618,1253 'must':935,969,2655 'n':2797,2800,2803 'name':85,107,159,562,580,615,1162,1284,1562,2190,2232,2475,2476,2508,2535,2574,2626,2629,2701,2707 'narrat':910,996,1006,1169,1268,2511,2586,2652,2687,3125 'need':1180,1798,2051,2693 'negoti':870,872 'never':821 'new':1839,2472,2620 'next':993,1700,2056,2334,2348,2564 'nois':2820,2848 'none':3179 'norm':2030 'normalis':3008,3042 'note':113,498,1997,2201,2883,2895,2961,3035 'null':2377 'number':469,488,944 'object':439 'observ':716,732,1636,2216,2977,3050 'occur':1225,1817 'on-demand':16,308 'one':208,316,318,627,633,976,1216,2280,2445,2458,2609,2613,2923 'one-sent':2279 'opportun':1624,2457,2529,2616 'order':663,680,730,774,809,904 'org':595,2203,3006,3013,3022 'outbound':181 'outcom':991,1509,1535,2066 'output':477,828,897,2283,2759,3056 'overal':2271,2435 'pace':2070 'paid':884,885 'paragraph':975,1004,1323,1525,1688,1849,2036,2584 'pars':2703 'part':2837 'partner':2872,2892,2900 'partner-context':2871,2891 'partnership':2857,2867 'pattern':441,454,528,537,1349,1364,1368,1404,1416,1432,1445,1491,1515,2513,2540,2550,2808 'pattern-to-d':536,1415 'peopl':222,1087,1165 'per':320,2925,3081 'period':1305,2142 'persist':450,1766,3096 'person':1096,1712,1780,1795 'phase':1285,1287,1291,1406,1412,1420 'pick':632 'pictur':215 'pipelin':15,163,816,1063,2858,2868 'placehold':69,91,459,466,481 'plain':1745 'plateau':1342 'point':1250,2189,2681 'posit':2462 'postur':1098,1617 'power':2721 'pre':1070 'pre-histori':1069 'predict':1442 'prefix':470 'prep':26 'present':255,621,1237,2903 'previous':1215 'primari':431,2317 'prior':1078,2372 'process':1782 'produc':131,165,287,506,899,2291,2430,2730,2806,3103,3159 'profil':1863 'progress':220,393,1278,1385,1490,1671,1697,1824,2019 'progression/regression':406 'project':324,326 'prose':955,1005 'prospect':843,844,2876,2885 'protocol':460 'provid':78,572,582,1379,1457 'pull':73,1587 'pure':1537 'qualifi':858 'qualiti':227,1391,2076,2080,2199,2214,2523,2784,2814,2898,2965,3049,3106 'queri':125,329 'question':209 'quot':201,946,3086 'random':2560,2767 'rate':2020 'rather':2025 'ratio':2103 'raw':195,702,714,723,762,823,964,1585,2206,3088 're':1081,1715,2310,2696,2993 're-engag':1080,1714 're-interpret':2992 're-run':2309 're-trigg':2695 'read':185,278,293,337,651,659,1182,1547,1723,1892,2322,2799,2802,3138 'read-on':277 'readabl':834,909 'reader':1173,2985 'readi':874,877 'readm':549 'real':483,2975 'realist':2065 'reason':395,419,442,533,787,793,805,2754,2980,2996 'recommend':235 'reconnect':1076 'reconstruct':1190,1557,2971,3071 'record':3135 'reduc':1886 'refer':116,565 'referenc':1675 'referr':2878 'regress':1245,1967,1968,2471,2619 'relat':1281 'relationship':1079,1705 'relev':3038 'reliabl':1358 'render':2326,2670 'report':1428,2997,3021 'request':41,2940 'requir':1792 'resolut':453,556,658,818 'resolv':830,1789,1801,1810,1919 'rest':2180,2243 'result':128 'return':126,2376,2935 'right':1122,2454,3005 'risk':1622,1706,1873,2444,2527,2612 'role':418,1091,1567,1576,1631,1658,1662,2241 'row':695,3121 'rs':764 'rs.deal':770 'rs.id':766 'rule':820 'run':302,317,2311,2368,2926,3063,3143 'sal':849,850 'sale':851,857,931,1049 'say':640,1595,2235 'sc':759 'sc.classification':751 'sc.classified':776 'sc.confidence':752 'sc.dimension':750 'sc.evidence':753 'sc.framework':748 'sc.id':745 'sc.signal':746,767 'schedul':305 'schema':79 'score':248,262 'search':583 'section':902,968,999,1030,1184,1186,1359,1427,1447,1452,1542,1719,1888,2073,2176,2223,2438,2483,2588,2591,2594,2597,2600,2603,2653,2663,2805,3052,3093,3107,3126,3161,3171 'see':80,547,2830 'seed':3009,3068 'select':346,359,372,588,670,708,744,791,2341 'sell':2882 'senior':919 'sentenc':979,2281,2446,2459,2610,2614 'sequenc':652,1040 'serv':2306,2684 'set':2401 'shift':1263,1953 'show':1996,3198 'sign':1494 'signal':196,394,400,415,703,712,724,742,757,763,939,1083,1234,1586,1677,1758,2115,2122,2248,2463,2536,2632,2798,2887,2913,2951,2973,3089,3152 'signific':1106 'silent':634,3041 'simultan':1256 'singl':9,45,175,1879,2187,2246 'single-d':44 'skill':60,97,115 'skill-deal-briefer' 'skip':1247 'snapshot':699 'someon':1016 'sourc':203,710,2116,2124,2127 'source-violetfleming47' 'spars':2165,3146,3187 'specif':64,937,1124,1164,1508,1779,1793,2192,2220,3078 'spent':1930 'sponsor':432 'sql':333,345,358,371,587,669,707,743,790,855,856,2340,2396,2490,2742 'stabl':1771,1987 'stage':87,402,594,817,825,836,839,842,845,848,854,860,863,868,873,880,883,886,891,894,1240,1243,1670,1701,1908,1911,1918,1924,1926,1933,1942,2207,2464,2504,2530,2548,2570,2642 'stage-by-stag':1923 'stakehold':417,425,437,1386,1544,1558,1640,1661,1691,2239,2515,2532,2622,2810,3092 'stale':895,896,2557,2948,2962,2972 'stale/archived':2956 'stall':1344,1496,1875,2470,2618 'stand':1121 'starter':112,123,496,502,1366,1377 'state':143,189,603,674,693,985,1196,1295,1549,1682,1726,1894,1915,1972,2046,2101,2501,2567,2770,2960,3033,3154 'status':356,369,382,1580 'step':664,700,735,779,2091,2119,2330,2388,2417,2485 'still':3181 'storag':2284 'store':2295 'stori':1011,1037,1151,1279 'strength':433,1438 'strong':1458,1693,2098,2262,2274,2440,2606 'strongest':2461 'structur':101,898,1692,2420,3128 'studi':923 'subsect':1572,1738 'substant':2161 'suffici':3173 'suggest':1532,1868 'summari':292,754,1027,2777 'summaris':285,2678 'supersed':2390,2405 'support':1599,1696,2197 'surfac':238 'synopsi':592,611,2510,2578,2580,2710,2713 'synthesis':1276 'system':497,520,553,711,1367,1378,1399,1425,2006,2125 'tabl':266,284,301,2324,2665 'taxonomi':401,426,449 'team':2303,2690 'technic':864,866 'tell':1008,1102,1328,1618 'test':1518 'text':2631,2660,2790 'thin':2174,3115,3205 'thing':1882,2450,2827 'thread':1655 'three':511,2945 'threshold':491 'throughout':959 'tier':719 'time':1022,1304,1906,1929,2141 'time-in-stag':1905 'timelin':412,1189,1317,1327,1450,1977,2151,2468,2512,2807,2970 'titl':1548,1575,1681 'told':1038 'topic-agent-skills' 'topic-agentic-ai' 'topic-agentic-workflow' 'topic-ai-agents' 'topic-ai-agents-framework' 'topic-b2b-sales-automation' 'topic-crm' 'topic-intelligence' 'topic-revops' 'topic-revops-automation' 'topic-sales-analysis' 'topic-sales-ops' 'total':1969 'toward':1099 'traceabl':273,781,802,2728,2740,2746,2981,3132 'track':2041 'trail':788 'trajectori':1111,1262,1334,1462,1531,1603,2024,2061 'transcript':77 'transit':403,1207,1244,1943 'treat':480,487 'trend':1983 'trigger':20,558,1229,1231,1233,1838,2697 'true':607,2416,2648 'truncat':2676 'trust':2112 'two':978 'type':448,596,713,1232,1734,2204,2750,3007,3014,3023 'typic':1489 'uncalibr':494 'unchang':2072 'unclassifi':1672 'underpin':2083 'understand':1176 'univers':103 'unresolv':1843,1855 'updat':2397 'upfront':3158 'upstream':2978 'use':135,386,390,398,422,445,510,574,1160,2716,2943,3024 'uuid':2561,2768,2773 'valid':1226,1938 'valu':65,598,1958,1962,2252,2506,2558,2572,2765,3015,3026 'valuabl':3195 'veloc':522,545,1388,1890,1893,1914,1927,1971,1998,2010,2520,2812 'verbatim':200,945,1581,3085 'version':599,667,682,688,943,1197,1200,1206,1212,1251,1272,1274,1301,1348,1375,1552,1605,1729,1898,1902,1936,1951,1964,2133,2137,2145,2336,2344,2349,2361,2386,2499,2502,2543,2565,2568,2637,2772,2795,3073,3082,3098,3150 'version-by-vers':1271 'via':330 'walk':1191 'want':1513 'weak':2100,2183,2196,2229,2276,2442,2608 'weak-confid':2182,2228 'well':3067 'well-seed':3066 'whether':1065,1238,1819,3002 'whole':2680 'win/loss/friction/velocity':1431 'without':1179,1444,2007,2308,2691 'wors':1770 'would':1044,1504,1511,1791,2050 'wrap':67 'write':263,916,1000,1147,1569,1735,2285,2487,2734,2737,3116,3130,3182 'written':1041","prices":[{"id":"22f80eae-bca0-46c5-89e4-f242f325e903","listingId":"9fcd9ebf-07a8-4401-a46a-089c1a8d18d9","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"violetfleming47","category":"deal-intelligence","install_from":"skills.sh"},"createdAt":"2026-05-14T07:05:43.660Z"}],"sources":[{"listingId":"9fcd9ebf-07a8-4401-a46a-089c1a8d18d9","source":"github","sourceId":"violetfleming47/deal-intelligence/deal-briefer","sourceUrl":"https://github.com/violetfleming47/deal-intelligence/tree/main/skills/deal-briefer","isPrimary":false,"firstSeenAt":"2026-05-14T07:05:43.660Z","lastSeenAt":"2026-05-18T19:06:27.019Z"}],"details":{"listingId":"9fcd9ebf-07a8-4401-a46a-089c1a8d18d9","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"violetfleming47","slug":"deal-briefer","github":{"repo":"violetfleming47/deal-intelligence","stars":15,"topics":["agent-skills","agentic-ai","agentic-workflow","ai-agents","ai-agents-framework","b2b-sales-automation","crm","intelligence","revops","revops-automation","sales-analysis","sales-ops"],"license":"mit","html_url":"https://github.com/violetfleming47/deal-intelligence","pushed_at":"2026-05-10T19:50:43Z","description":"Turn CRM and transcript data into structured, versioned deal intelligence. 6 agents, 3 frameworks, harness-agnostic. Works with any CRM, any transcript provider, any LLM.","skill_md_sha":"dab7d99082171f2caaae111ea3cf80d821b5a712","skill_md_path":"skills/deal-briefer/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/violetfleming47/deal-intelligence/tree/main/skills/deal-briefer"},"layout":"multi","source":"github","category":"deal-intelligence","frontmatter":{"name":"deal-briefer","description":"Forensic deep dive on a single deal from the Deal Intelligence pipeline. On-demand only. Triggered by 'deep dive on [deal]', 'prep me for the [company] call', 'analyse [deal]', 'what's happening with [company]', or any request for a single-deal analytical deep dive backed by evidence from di_* tables."},"skills_sh_url":"https://skills.sh/violetfleming47/deal-intelligence/deal-briefer"},"updatedAt":"2026-05-18T19:06:27.019Z"}}