feat: add Negotiation Squad v1.0.0 — 8 elite negotiation experts#16
Open
rogowski wants to merge 1 commit intoSynkraAI:mainfrom
Open
feat: add Negotiation Squad v1.0.0 — 8 elite negotiation experts#16rogowski wants to merge 1 commit intoSynkraAI:mainfrom
rogowski wants to merge 1 commit intoSynkraAI:mainfrom
Conversation
…10 tasks, 2 workflows) Full sales negotiation pipeline with research-backed frameworks: - Robert Cialdini (7 Principles of Influence) for buyer profiling - Chet Holmes (Dream 100) for ideal client identification - Jeb Blount (Fanatical Prospecting) for multi-channel outreach - Oren Klaff (STRONG Method) for pitch creation with frame control - Neil Rackham (SPIN Selling, 35K+ calls research) for discovery - Chris Voss (FBI Tactical Empathy) for negotiation - Jim Camp (Start With No) for strategic closing - William Ury (Harvard Principled Negotiation) for strategic advisory Includes 3 templates, 2 checklists, and 2 workflows (full pipeline + focused negotiation). Co-Authored-By: Claude Opus 4.6 (1M context) <[email protected]>
📝 WalkthroughWalkthroughA new Negotiation Squad system is introduced with eight specialized agents (Chet Holmes, Chris Voss, Jeb Blount, Jim Camp, Neil Rackham, Oren Klaff, Robert Cialdini, William Ury), ten task definitions, two workflows, two checklists, three templates, and supporting configuration files. All changes are declarative, providing structured negotiation frameworks and agent specifications without introducing executable code. Changes
Estimated code review effort🎯 4 (Complex) | ⏱️ ~50 minutes
🚥 Pre-merge checks | ✅ 3✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
📝 Coding Plan
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment Tip You can enable review details to help with troubleshooting, context usage and more.Enable the |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 18
🧹 Nitpick comments (4)
squads/negotiation/tasks/plan-negotiation.md (1)
18-23: Clarify minimum required inputs for workflow-invoked runs.In the fixed
wf-negotiate-dealflow, this step gets only phase_1 output by default; making required-vs-optional inputs explicit here would reduce ambiguity and improve run consistency.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@squads/negotiation/tasks/plan-negotiation.md` around lines 18 - 23, The step currently lists possible inputs but doesn't state which are required when the wf-negotiate-deal flow invokes this step with only phase_1 output by default; update the "input:" block in plan-negotiation.md to explicitly mark required vs optional inputs (e.g., required: "Deal context and background" and "phase_1 output" or similar) and add a short note referencing wf-negotiate-deal and phase_1 so callers know the minimum fields the workflow will supply and which additional fields are optional for manual or extended runs.squads/negotiation/workflows/wf-full-pipeline.yaml (1)
108-114: Clarify execution mode semantics inflow_rules.“Phases are sequential” and “can be entered at any point” conflict semantically. Consider explicitly defining two modes (strict sequential vs partial start-at-phase) to avoid orchestration ambiguity.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@squads/negotiation/workflows/wf-full-pipeline.yaml` around lines 108 - 114, The flow_rules text is semantically ambiguous about phase ordering; update the workflow definition (flow_rules) to explicitly introduce an execution_mode field with two modes: "sequential" (enforce phases run strictly in order and require each checkpoint to pass before next) and "start_at_phase" (allow starting at a specified phase but still require prior-phase checkpoints to be validated or marked as waived), and document how Phase 0 (Define Offer) and Phase 1 (Profile Buyer) are treated under each mode plus how the Strategic Advisor invocation and partial runs (e.g., Phase 6-7) interact with checkpoint validation; modify any references in the pipeline metadata to use execution_mode and ensure checkpoint validation logic and orchestration will consult execution_mode when deciding whether to permit phase entry or require fixes.squads/negotiation/agents/jeb-blount.md (1)
19-20: Unify tier value with squad manifest entry.This file declares tier
2, while the squad manifest registersjeb-blountas tier1. Aligning these prevents inconsistent orchestration behavior.Also applies to: 85-86
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@squads/negotiation/agents/jeb-blount.md` around lines 19 - 20, Update the tier value in this agent metadata so it matches the squad manifest: change the YAML field "tier" in squads/negotiation/agents/jeb-blount.md from 2 to 1, and also fix the duplicate occurrences referenced around lines 85-86 to ensure all "tier" entries for jeb-blount are consistent with the squad manifest's tier 1.squads/negotiation/agents/oren-klaff.md (1)
19-20: Align agent tier with squad registration to avoid routing ambiguity.This file marks Oren as tier
2, while the squad manifest registersoren-klaffas tier1. Keep a single source of truth for tier to prevent orchestration inconsistencies.Also applies to: 83-84
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@squads/negotiation/agents/oren-klaff.md` around lines 19 - 20, The agent metadata in the Oren Klaff markdown has a tier mismatch: update the tier value in the frontmatter for the agent "oren-klaff" (found in this file's top matter and also at the duplicate block around lines referenced as 83-84) so it matches the squad manifest's registered tier (change tier: 2 to tier: 1 or vice versa depending on the squad manifest); ensure both occurrences in this file use the same tier number to keep a single source of truth and avoid routing ambiguity.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@squads/negotiation/agents/chris-voss.md`:
- Around line 28-43: The command "*chat-mode" is referenced but not defined in
the command_loader, which will cause runtime lookup failures; add an entry for
"*chat-mode" under the command_loader map (similar to other entries like
"*negotiate-deal" and "*plan-negotiation") with an appropriate description and
any required/optional task/template references so the command lookup succeeds;
also search the file for other missing command entries mentioned (the instance
referenced around the later block) and add matching command_loader entries for
each missing symbol to ensure all declared commands are registered.
In `@squads/negotiation/agents/jeb-blount.md`:
- Around line 66-68: The commands "*phone-script" and "*pipeline-math"
referenced in the command_loader mapping and in the dependencies are unresolved
because their task files are missing; fix this by either (A) adding the
corresponding task markdown files named phone-script.md and pipeline-math.md
into the negotiation tasks directory with the implementations expected by the
command_loader, or (B) remove the "*phone-script" and "*pipeline-math" entries
from the command mapping and from the dependency list so there are no dangling
command references; update the command_loader mapping and dependency list (where
the two command symbols appear) accordingly.
In `@squads/negotiation/agents/negotiation-chief.md`:
- Around line 44-107: The command_loader is missing an entry for "*chat-mode"
required by CRITICAL_LOADER_RULE; add a new command_loader entry named
"*chat-mode" (e.g., alongside "*help" and "*exit") with a brief description like
"Enter chat mode" and an empty requires list (requires: []) so
command-resolution won't fail—update the command_loader map that contains
entries such as "*help", "*exit", "*plan-negotiation" to include "*chat-mode".
- Around line 75-99: The focused workflow wf-negotiate-deal is missing from the
chief’s command routing: add a runnable command entry for "*negotiate-deal" that
exposes the wf-negotiate-deal.yaml workflow (similar to how "*full-pipeline" is
exposed), include its required dependency (wf-negotiate-deal.yaml) and any
relevant optional templates/checklists, and replicate the same fix for the other
occurrence of the mapping elsewhere in the file so the focused workflow is
reachable via the chief’s command interface.
- Around line 156-169: The copy currently says "7 elite" in multiple places but
the squad roster defines 8 specialist experts plus the orchestrator; update all
instances—specifically the header string "1.0: Initial creation with 7 elite
mind agents", the persona.identity line "I orchestrate a team of 7 elite
negotiation minds", and the background line "7 elite minds spanning..."—to
reflect the correct count and consistent phrasing (e.g., "8 elite minds" or "8
specialist experts plus the orchestrator") so the text matches the actual squad
composition.
In `@squads/negotiation/agents/neil-rackham.md`:
- Around line 33-38: The squad command manifest contains command tasks
"*qualify-opportunity" and "*objection-prevention" in agents/neil-rackham.md but
the squad task registry is missing matching entries, causing unresolved task
loading; update the squad task manifest to include corresponding task IDs and
file references for qualify-opportunity and objection-prevention (and any other
missing task IDs noted around the other occurrences at the same locations) by
adding the exact task IDs and their task file paths to the squad task list so
the loader can resolve them (ensure names match the symbols
"*qualify-opportunity" and "*objection-prevention" exactly).
- Around line 28-43: The command_loader mapping is missing an entry for the
"*chat-mode" command referenced at lines 276-279; add a "*chat-mode" key under
command_loader (matching the style of "*help" and "*exit") and supply an
appropriate requires/optional list (e.g., requires: [] and optional: [] or any
task/template files needed by chat-mode) so the loader lookup rules will find
and load the command; update the command_loader block that contains
"*spin-discovery", "*qualify-opportunity", etc., to include "*chat-mode" with
the correct requires/optional values.
In `@squads/negotiation/agents/oren-klaff.md`:
- Around line 64-67: Mapping entries "*frame-strategy" and "*status-alignment"
in oren-klaff.md point to non-existent task files and will break at runtime;
either create the missing task documents named frame-strategy.md and
status-alignment.md under the negotiation tasks directory with the expected task
content, or update the mappings in the command map to reference the correct
existing task files (e.g., replace the values for "*frame-strategy" and
"*status-alignment" with the proper filenames). Locate the mapping block
containing "*create-pitch", "*frame-strategy", "*status-alignment", and "*help"
to apply the fix.
In `@squads/negotiation/agents/robert-cialdini.md`:
- Around line 28-41: The command_loader mapping is missing an entry for
"*chat-mode" causing runtime command-resolution failures; add a "*chat-mode" key
under command_loader (similar to other entries like "*profile-buyer" and
"*influence-strategy") with an appropriate description and required/optional
task/template references so the runtime can load the chat-mode command; ensure
you also add the same "*chat-mode" mapping to the other duplicated sections
noted (the blocks around the other occurrences) so all command_loader blocks
consistently include "*chat-mode".
In `@squads/negotiation/agents/william-ury.md`:
- Around line 28-43: The command "*chat-mode" is referenced elsewhere but
missing from the command_loader mapping, which breaks resolution under
CRITICAL_LOADER_RULE; add an entry for "*chat-mode" to the command_loader block
(mirroring the pattern used by "*help" and "*exit") including appropriate
description and required/optional task/template keys, and ensure any occurrences
noted at lines 45-48 and 277-280 are consistent with the new command_loader
registration so CRITICAL_LOADER_RULE can resolve "*chat-mode" correctly.
- Around line 33-38: The entries "*analyze-batna" and "*breakthrough-strategy"
reference missing task files (analyze-batna.md and breakthrough-strategy.md)
causing runtime failures; fix by either adding those task files with the
expected task content and metadata so the "requires" arrays resolve, or
remove/replace these command entries and any invocations that reference
"*analyze-batna" or "*breakthrough-strategy"; if you replace them, update the
"requires" arrays on the command definitions (the two starred command keys) to
point to existing task filenames instead.
In `@squads/negotiation/config.yaml`:
- Around line 22-43: The pipeline_flow is missing the initial "Define Offer"
phase which must be added before "buyer-profiling"; update the pipeline_flow
array to insert an entry with phase: "define-offer", agent: "alex-hormozi" (or
the appropriate agent for offer design) and a brief description like "Create and
validate core offer, pricing, and positioning" so downstream orchestration that
expects the full pipeline order (define-offer → buyer-profiling → ...) remains
consistent; ensure the new entry is placed at the top of pipeline_flow and
follows the same YAML formatting as the other phases.
- Around line 85-97: The task_standards required_fields list (under
task_standards, validator "NG-TP-001") is out of sync with the actual
negotiation task schema: config expects task_name, responsible_executor, and
status while the PR tasks use name and executor and omit status; either update
the config to match the actual schema (replace task_name with name,
responsible_executor with executor, and make status optional/remove it) or
update every negotiation task file to use the canonical fields (rename
name→task_name, executor→responsible_executor and add a status field) so that
enforce_anatomy: true validation passes; pick one approach and apply
consistently across the config (required_fields) and the task files referenced
by NG-TP-001.
In `@squads/negotiation/README.md`:
- Around line 32-44: The README has markdownlint MD040/MD058 issues from
unlabeled code fences and missing blank lines around tables; update every fenced
block (e.g., the pipeline block and later code blocks around "Agents (9)" and
"Tasks (10)") to use explicit fence languages like ```text or ```bash and ensure
there is a blank line before and after each table and code fence (add or
maintain a separating blank line between headings, code fences, and tables).
Locate and fix the unlabeled fences in the negotiation README sections (the
pipeline example, the Agents and Tasks sections, and other ranges noted) so all
fences include a language token and tables have proper surrounding spacing to
satisfy MD040/MD058.
- Around line 5-7: Update the Portuguese user-facing copy to normalize
orthography and add missing diacritics: replace "analise" with "análise", "ate"
with "até", "negociacao" with "negociação", "psicologico" with "psicológico",
and "estrategico" with "estratégico"; review the README content (including the
other occurrences noted at lines 11-14) and apply the same diacritic corrections
so all user-facing phrases like "Squad de 8 elite minds reais para o pipeline
completo de negociacao profissional" become properly accented and natural in
Portuguese.
In `@squads/negotiation/tasks/profile-buyer.md`:
- Around line 80-110: The output_format block in this file diverges from the
canonical buyer-profile-tmpl schema; update the sections to exactly match
buyer-profile-tmpl by replacing or renaming the current headers and fields
(e.g., change "Decision Role" to "Decision Profile", add "Authority Signals to
Demonstrate", and ensure "Social Proof to Deploy" matches the template wording
and order), preserve the existing placeholder tokens (e.g., {name}, {title},
{company}, {top principle}) and table structure but align column/section names
and section ordering with buyer-profile-tmpl so downstream consumers see the
identical contract.
In `@squads/negotiation/templates/proposal-tmpl.md`:
- Around line 41-44: The Timeline table starts immediately after the "###
Timeline" heading which violates MD058; fix by inserting a single blank line
between the "### Timeline" heading and the table (the table beginning with "|
Phase | Duration | Deliverable |") so the heading and the table are separated by
an empty line.
In `@squads/negotiation/workflows/wf-negotiate-deal.yaml`:
- Around line 31-39: phase_3 currently only lists input_from: ["phase_1",
"phase_2"] but the task tasks/negotiate-deal.md needs explicit structured inputs
(Deal context, Current terms / proposal, Known objections); update phase_3 to
supply those required fields by either expanding input_from to reference the
specific producers (e.g., ["phase_1.deal_context", "phase_2.current_terms",
"phase_2.known_objections"]) or add an inputs section that maps Deal context,
Current terms / proposal, and Known objections to the appropriate upstream
outputs so tasks/negotiate-deal.md always receives the required structured
context.
---
Nitpick comments:
In `@squads/negotiation/agents/jeb-blount.md`:
- Around line 19-20: Update the tier value in this agent metadata so it matches
the squad manifest: change the YAML field "tier" in
squads/negotiation/agents/jeb-blount.md from 2 to 1, and also fix the duplicate
occurrences referenced around lines 85-86 to ensure all "tier" entries for
jeb-blount are consistent with the squad manifest's tier 1.
In `@squads/negotiation/agents/oren-klaff.md`:
- Around line 19-20: The agent metadata in the Oren Klaff markdown has a tier
mismatch: update the tier value in the frontmatter for the agent "oren-klaff"
(found in this file's top matter and also at the duplicate block around lines
referenced as 83-84) so it matches the squad manifest's registered tier (change
tier: 2 to tier: 1 or vice versa depending on the squad manifest); ensure both
occurrences in this file use the same tier number to keep a single source of
truth and avoid routing ambiguity.
In `@squads/negotiation/tasks/plan-negotiation.md`:
- Around line 18-23: The step currently lists possible inputs but doesn't state
which are required when the wf-negotiate-deal flow invokes this step with only
phase_1 output by default; update the "input:" block in plan-negotiation.md to
explicitly mark required vs optional inputs (e.g., required: "Deal context and
background" and "phase_1 output" or similar) and add a short note referencing
wf-negotiate-deal and phase_1 so callers know the minimum fields the workflow
will supply and which additional fields are optional for manual or extended
runs.
In `@squads/negotiation/workflows/wf-full-pipeline.yaml`:
- Around line 108-114: The flow_rules text is semantically ambiguous about phase
ordering; update the workflow definition (flow_rules) to explicitly introduce an
execution_mode field with two modes: "sequential" (enforce phases run strictly
in order and require each checkpoint to pass before next) and "start_at_phase"
(allow starting at a specified phase but still require prior-phase checkpoints
to be validated or marked as waived), and document how Phase 0 (Define Offer)
and Phase 1 (Profile Buyer) are treated under each mode plus how the Strategic
Advisor invocation and partial runs (e.g., Phase 6-7) interact with checkpoint
validation; modify any references in the pipeline metadata to use execution_mode
and ensure checkpoint validation logic and orchestration will consult
execution_mode when deciding whether to permit phase entry or require fixes.
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: de624a72-080a-4769-bb65-bbc204340883
📒 Files selected for processing (29)
squads/negotiation/README.mdsquads/negotiation/agents/chet-holmes.mdsquads/negotiation/agents/chris-voss.mdsquads/negotiation/agents/jeb-blount.mdsquads/negotiation/agents/jim-camp.mdsquads/negotiation/agents/negotiation-chief.mdsquads/negotiation/agents/neil-rackham.mdsquads/negotiation/agents/oren-klaff.mdsquads/negotiation/agents/robert-cialdini.mdsquads/negotiation/agents/william-ury.mdsquads/negotiation/checklists/deal-readiness.mdsquads/negotiation/checklists/negotiation-prep.mdsquads/negotiation/config.yamlsquads/negotiation/squad.yamlsquads/negotiation/tasks/close-deal.mdsquads/negotiation/tasks/create-pitch.mdsquads/negotiation/tasks/define-offer.mdsquads/negotiation/tasks/handle-objection.mdsquads/negotiation/tasks/identify-dream-clients.mdsquads/negotiation/tasks/negotiate-deal.mdsquads/negotiation/tasks/plan-negotiation.mdsquads/negotiation/tasks/profile-buyer.mdsquads/negotiation/tasks/prospect-outreach.mdsquads/negotiation/tasks/spin-discovery.mdsquads/negotiation/templates/buyer-profile-tmpl.mdsquads/negotiation/templates/negotiation-plan-tmpl.mdsquads/negotiation/templates/proposal-tmpl.mdsquads/negotiation/workflows/wf-full-pipeline.yamlsquads/negotiation/workflows/wf-negotiate-deal.yaml
Comment on lines
+28
to
+43
| command_loader: | ||
| "*negotiate-deal": | ||
| description: "Create tactical negotiation playbook" | ||
| requires: ["tasks/negotiate-deal.md"] | ||
| optional: ["templates/negotiation-plan-tmpl.md"] | ||
| "*handle-objection": | ||
| description: "Handle specific objection" | ||
| requires: ["tasks/handle-objection.md"] | ||
| "*plan-negotiation": | ||
| description: "Full negotiation strategy" | ||
| requires: ["tasks/plan-negotiation.md"] | ||
| "*help": | ||
| requires: [] | ||
| "*exit": | ||
| requires: [] | ||
|
|
There was a problem hiding this comment.
*chat-mode is declared as a command but missing from command_loader.
With the rule on Line 45-48 (“ANY command”), invoking *chat-mode can fail lookup at runtime.
✅ Minimal fix
command_loader:
@@
"*help":
requires: []
+ "*chat-mode":
+ requires: []
"*exit":
requires: []Also applies to: 228-231
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@squads/negotiation/agents/chris-voss.md` around lines 28 - 43, The command
"*chat-mode" is referenced but not defined in the command_loader, which will
cause runtime lookup failures; add an entry for "*chat-mode" under the
command_loader map (similar to other entries like "*negotiate-deal" and
"*plan-negotiation") with an appropriate description and any required/optional
task/template references so the command lookup succeeds; also search the file
for other missing command entries mentioned (the instance referenced around the
later block) and add matching command_loader entries for each missing symbol to
ensure all declared commands are registered.
Comment on lines
+66
to
+68
| "*prospect-outreach" : "tasks/prospect-outreach.md" | ||
| "*phone-script" : "tasks/phone-script.md" | ||
| "*pipeline-math" : "tasks/pipeline-math.md" |
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash
set -euo pipefail
for f in \
squads/negotiation/tasks/prospect-outreach.md \
squads/negotiation/tasks/phone-script.md \
squads/negotiation/tasks/pipeline-math.md
do
if [ -f "$f" ]; then
echo "OK: $f"
else
echo "MISSING: $f"
fi
doneRepository: SynkraAI/aiox-squads
Length of output: 214
🏁 Script executed:
cat -n squads/negotiation/agents/jeb-blount.md | sed -n '60,75p'Repository: SynkraAI/aiox-squads
Length of output: 928
🏁 Script executed:
cat -n squads/negotiation/agents/jeb-blount.md | sed -n '255,270p'Repository: SynkraAI/aiox-squads
Length of output: 747
🏁 Script executed:
wc -l squads/negotiation/agents/jeb-blount.mdRepository: SynkraAI/aiox-squads
Length of output: 107
🏁 Script executed:
fd -name "phone-script.md" -o -name "pipeline-math.md"Repository: SynkraAI/aiox-squads
Length of output: 294
🏁 Script executed:
find . -name "*.md" -path "*/negotiation/tasks/*" -type fRepository: SynkraAI/aiox-squads
Length of output: 519
🏁 Script executed:
fd "phone-script"Repository: SynkraAI/aiox-squads
Length of output: 46
🏁 Script executed:
fd "pipeline-math"Repository: SynkraAI/aiox-squads
Length of output: 46
Command targets are unresolved for two public commands.
*phone-script and *pipeline-math are mapped in the command_loader (lines 66-68) and listed in dependencies (lines 263-265), but their corresponding task files—squads/negotiation/tasks/phone-script.md and squads/negotiation/tasks/pipeline-math.md—do not exist in the repository. Command execution will fail when users invoke these commands. Create the missing files or remove the unimplemented command mappings.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@squads/negotiation/agents/jeb-blount.md` around lines 66 - 68, The commands
"*phone-script" and "*pipeline-math" referenced in the command_loader mapping
and in the dependencies are unresolved because their task files are missing; fix
this by either (A) adding the corresponding task markdown files named
phone-script.md and pipeline-math.md into the negotiation tasks directory with
the implementations expected by the command_loader, or (B) remove the
"*phone-script" and "*pipeline-math" entries from the command mapping and from
the dependency list so there are no dangling command references; update the
command_loader mapping and dependency list (where the two command symbols
appear) accordingly.
Comment on lines
+44
to
+107
| command_loader: | ||
| "*define-offer": | ||
| description: "Define your product/service offer for negotiation" | ||
| requires: ["tasks/define-offer.md"] | ||
| optional: ["templates/proposal-tmpl.md"] | ||
|
|
||
| "*profile-buyer": | ||
| description: "Analyze buyer psychology using Cialdini's 7 Principles" | ||
| requires: ["tasks/profile-buyer.md"] | ||
| optional: ["templates/buyer-profile-tmpl.md"] | ||
|
|
||
| "*identify-dream-clients": | ||
| description: "Identify Dream 100 ideal clients using Chet Holmes strategy" | ||
| requires: ["tasks/identify-dream-clients.md"] | ||
| optional: [] | ||
|
|
||
| "*prospect-outreach": | ||
| description: "Execute multi-channel prospecting using Jeb Blount framework" | ||
| requires: ["tasks/prospect-outreach.md"] | ||
| optional: [] | ||
|
|
||
| "*create-pitch": | ||
| description: "Create pitch using Oren Klaff STRONG method" | ||
| requires: ["tasks/create-pitch.md"] | ||
| optional: [] | ||
|
|
||
| "*spin-discovery": | ||
| description: "Run SPIN Selling discovery using Neil Rackham framework" | ||
| requires: ["tasks/spin-discovery.md"] | ||
| optional: [] | ||
|
|
||
| "*negotiate-deal": | ||
| description: "Negotiate deal using Chris Voss tactical empathy" | ||
| requires: ["tasks/negotiate-deal.md"] | ||
| optional: ["templates/negotiation-plan-tmpl.md"] | ||
|
|
||
| "*close-deal": | ||
| description: "Close deal using Jim Camp Start With No system" | ||
| requires: ["tasks/close-deal.md"] | ||
| optional: [] | ||
|
|
||
| "*plan-negotiation": | ||
| description: "Create complete negotiation plan using William Ury + Voss" | ||
| requires: ["tasks/plan-negotiation.md"] | ||
| optional: ["templates/negotiation-plan-tmpl.md", "checklists/negotiation-prep.md"] | ||
|
|
||
| "*handle-objection": | ||
| description: "Handle specific objection using combined frameworks" | ||
| requires: ["tasks/handle-objection.md"] | ||
| optional: [] | ||
|
|
||
| "*full-pipeline": | ||
| description: "Run full negotiation pipeline end-to-end" | ||
| requires: ["workflows/wf-full-pipeline.yaml"] | ||
| optional: ["checklists/deal-readiness.md"] | ||
|
|
||
| "*help": | ||
| description: "Show available commands" | ||
| requires: [] | ||
|
|
||
| "*exit": | ||
| description: "Exit agent" | ||
| requires: [] | ||
|
|
There was a problem hiding this comment.
chat-mode is missing from command_loader.
Line 335-338 defines chat-mode, but CRITICAL_LOADER_RULE requires every command to exist in command_loader. Add *chat-mode with empty requirements to avoid command-resolution failures.
✅ Minimal fix
"*help":
description: "Show available commands"
requires: []
+ "*chat-mode":
+ description: "Open conversation about negotiation strategy"
+ requires: []
"*exit":
description: "Exit agent"
requires: []Also applies to: 335-338
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@squads/negotiation/agents/negotiation-chief.md` around lines 44 - 107, The
command_loader is missing an entry for "*chat-mode" required by
CRITICAL_LOADER_RULE; add a new command_loader entry named "*chat-mode" (e.g.,
alongside "*help" and "*exit") with a brief description like "Enter chat mode"
and an empty requires list (requires: []) so command-resolution won't
fail—update the command_loader map that contains entries such as "*help",
"*exit", "*plan-negotiation" to include "*chat-mode".
Comment on lines
+75
to
+99
| "*negotiate-deal": | ||
| description: "Negotiate deal using Chris Voss tactical empathy" | ||
| requires: ["tasks/negotiate-deal.md"] | ||
| optional: ["templates/negotiation-plan-tmpl.md"] | ||
|
|
||
| "*close-deal": | ||
| description: "Close deal using Jim Camp Start With No system" | ||
| requires: ["tasks/close-deal.md"] | ||
| optional: [] | ||
|
|
||
| "*plan-negotiation": | ||
| description: "Create complete negotiation plan using William Ury + Voss" | ||
| requires: ["tasks/plan-negotiation.md"] | ||
| optional: ["templates/negotiation-plan-tmpl.md", "checklists/negotiation-prep.md"] | ||
|
|
||
| "*handle-objection": | ||
| description: "Handle specific objection using combined frameworks" | ||
| requires: ["tasks/handle-objection.md"] | ||
| optional: [] | ||
|
|
||
| "*full-pipeline": | ||
| description: "Run full negotiation pipeline end-to-end" | ||
| requires: ["workflows/wf-full-pipeline.yaml"] | ||
| optional: ["checklists/deal-readiness.md"] | ||
|
|
There was a problem hiding this comment.
Focused workflow wf-negotiate-deal is not exposed as a runnable command.
You include wf-negotiate-deal.yaml in dependencies (Line 138), but command routing exposes only *full-pipeline for workflows. This makes the focused workflow effectively unreachable via the chief’s command interface.
💡 Suggested command exposure
command_loader:
@@
"*full-pipeline":
description: "Run full negotiation pipeline end-to-end"
requires: ["workflows/wf-full-pipeline.yaml"]
optional: ["checklists/deal-readiness.md"]
+ "*active-deal":
+ description: "Run focused active-deal negotiation workflow"
+ requires: ["workflows/wf-negotiate-deal.yaml"]
+ optional: ["checklists/negotiation-prep.md"]
commands:
@@
- name: full-pipeline
visibility: [full, quick]
description: "Run complete deal pipeline end-to-end"
loader: "workflows/wf-full-pipeline.yaml"
+ - name: active-deal
+ visibility: [full, quick]
+ description: "Run focused active-deal negotiation workflow"
+ loader: "workflows/wf-negotiate-deal.yaml"Also applies to: 330-334
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@squads/negotiation/agents/negotiation-chief.md` around lines 75 - 99, The
focused workflow wf-negotiate-deal is missing from the chief’s command routing:
add a runnable command entry for "*negotiate-deal" that exposes the
wf-negotiate-deal.yaml workflow (similar to how "*full-pipeline" is exposed),
include its required dependency (wf-negotiate-deal.yaml) and any relevant
optional templates/checklists, and replicate the same fix for the other
occurrence of the mapping elsewhere in the file so the focused workflow is
reachable via the chief’s command interface.
Comment on lines
+156
to
+169
| - "1.0: Initial creation with 7 elite mind agents" | ||
|
|
||
| persona: | ||
| role: "Master Negotiation Orchestrator & Deal Strategist" | ||
| style: "Strategic, analytical, decisive — routes to the right expert for each phase" | ||
| identity: | | ||
| I'm the Deal Architect — I orchestrate a team of 7 elite negotiation minds, | ||
| each with documented frameworks and real-world results. I diagnose your situation, | ||
| route you to the right expert, and coordinate the full pipeline from buyer | ||
| profiling to deal closing. | ||
| focus: "Maximizing deal value through the right expert at the right phase" | ||
| background: | | ||
| Built on the combined frameworks of 7 elite minds spanning FBI hostage | ||
| negotiation, Harvard principled negotiation, neuroscience-based pitching, |
There was a problem hiding this comment.
Mind-count text is inconsistent with the actual squad composition.
Line 156 and Line 162 describe “7 elite minds,” but this squad defines 8 specialist experts plus the orchestrator. Please align the count in copy to avoid user confusion.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@squads/negotiation/agents/negotiation-chief.md` around lines 156 - 169, The
copy currently says "7 elite" in multiple places but the squad roster defines 8
specialist experts plus the orchestrator; update all instances—specifically the
header string "1.0: Initial creation with 7 elite mind agents", the
persona.identity line "I orchestrate a team of 7 elite negotiation minds", and
the background line "7 elite minds spanning..."—to reflect the correct count and
consistent phrasing (e.g., "8 elite minds" or "8 specialist experts plus the
orchestrator") so the text matches the actual squad composition.
Comment on lines
+5
to
+7
| Squad de 8 elite minds reais para o pipeline completo de negociacao profissional. | ||
| Desde a analise do perfil psicologico do comprador ate o fechamento estrategico, | ||
| cada agente tem um framework documentado, testado em milhares de negociacoes reais. |
There was a problem hiding this comment.
Normalize Portuguese orthography in user-facing copy.
There are PT-BR accent/paronym issues (e.g., “analise”/“análise”, “ate”/“até”) that reduce documentation quality.
Also applies to: 11-14
🧰 Tools
🪛 LanguageTool
[uncategorized] ~6-~6: Quando escrita sem acento, esta palavra é um verbo. Se pretende referir-se a um substantivo ou adjetivo, deve utilizar a forma acentuada.
Context: ...pleto de negociacao profissional. Desde a analise do perfil psicologico do comprador ate ...
(PARONYM_ANALISE_363)
[uncategorized] ~6-~6: Se pretende usar a preposição ‘até’, acrescente o acento agudo.
Context: ...lise do perfil psicologico do comprador ate o fechamento estrategico, cada agente t...
(CONFUSÃO_ATE_ATÉ)
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@squads/negotiation/README.md` around lines 5 - 7, Update the Portuguese
user-facing copy to normalize orthography and add missing diacritics: replace
"analise" with "análise", "ate" with "até", "negociacao" with "negociação",
"psicologico" with "psicológico", and "estrategico" with "estratégico"; review
the README content (including the other occurrences noted at lines 11-14) and
apply the same diacritic corrections so all user-facing phrases like "Squad de 8
elite minds reais para o pipeline completo de negociacao profissional" become
properly accented and natural in Portuguese.
Comment on lines
+32
to
+44
| ``` | ||
| Pipeline Phase Expert Framework | ||
| ───────────────────────────────────────────────────────────── | ||
| Profile Buyer → 🧠 Robert Cialdini → 7 Principles of Influence | ||
| Find Clients → 🎯 Chet Holmes → Dream 100 Strategy | ||
| Prospect → 📞 Jeb Blount → Fanatical Prospecting | ||
| Pitch → 🎬 Oren Klaff → STRONG Method + Frame Control | ||
| Discover → 🔍 Neil Rackham → SPIN Selling (35K+ calls research) | ||
| Negotiate → 🎯 Chris Voss → Tactical Empathy (FBI) | ||
| Close → ⚔️ Jim Camp → Start With No System | ||
| Advisor → 🕊️ William Ury → Principled Negotiation (Harvard) | ||
| Orchestrator → 🤝 Deal Architect → Routes to right expert per phase | ||
| ``` |
There was a problem hiding this comment.
Fix markdownlint violations (fence languages + table spacing).
Current fences and table spacing trigger MD040/MD058 warnings, which can fail doc lint pipelines.
Suggested doc-lint fix pattern
-```
+```text
Pipeline Phase Expert Framework
...
-```
+```
### Agents (9)
+
| Agent | Expert | Framework | Phase |
|-------|--------|-----------|-------|
...
+
### Tasks (10)
+
| Task | Purpose | Expert |
...-```
+```bash
/negotiation:negotiation-chief
-```
+```Also applies to: 49-93, 97-99, 102-106, 109-112, 115-119, 123-159
🧰 Tools
🪛 markdownlint-cli2 (0.21.0)
[warning] 32-32: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@squads/negotiation/README.md` around lines 32 - 44, The README has
markdownlint MD040/MD058 issues from unlabeled code fences and missing blank
lines around tables; update every fenced block (e.g., the pipeline block and
later code blocks around "Agents (9)" and "Tasks (10)") to use explicit fence
languages like ```text or ```bash and ensure there is a blank line before and
after each table and code fence (add or maintain a separating blank line between
headings, code fences, and tables). Locate and fix the unlabeled fences in the
negotiation README sections (the pipeline example, the Agents and Tasks
sections, and other ranges noted) so all fences include a language token and
tables have proper surrounding spacing to satisfy MD040/MD058.
Comment on lines
+80
to
+110
| output_format: | | ||
| ## Buyer DNA Profile: {buyer name} | ||
|
|
||
| ### Buyer Overview | ||
| - **Name:** {name} | ||
| - **Role:** {title} at {company} | ||
| - **Decision Role:** {decision-maker | influencer | gatekeeper} | ||
| - **Personality Type:** {analytical | expressive | driver | amiable} | ||
|
|
||
| ### Influence Lever Ranking | ||
| | Rank | Principle | Score | Recommended Action | | ||
| |------|-----------|-------|--------------------| | ||
| | 1 | {top principle} | {X}/10 | {specific action} | | ||
| | 2 | {2nd principle} | {X}/10 | {specific action} | | ||
| | 3 | {3rd principle} | {X}/10 | {specific action} | | ||
| | 4-7 | {others} | {X}/10 | {notes} | | ||
|
|
||
| ### Pre-Suasion Strategy | ||
| **Before the conversation:** | ||
| - {what to do to prime the buyer before the actual pitch} | ||
| - {environment, timing, and framing recommendations} | ||
|
|
||
| ### Approach Recommendations | ||
| - **Opening:** {how to open based on top influence levers} | ||
| - **Building Trust:** {specific trust signals to deploy} | ||
| - **Creating Urgency:** {scarcity/loss aversion if applicable} | ||
| - **Social Proof to Deploy:** {specific references, case studies, testimonials} | ||
|
|
||
| ### Red Flags / Avoid | ||
| - {what NOT to do with this buyer type} | ||
| - {principles that could backfire} |
There was a problem hiding this comment.
Align task output schema with buyer-profile-tmpl to avoid contract drift.
The output sections here diverge from squads/negotiation/templates/buyer-profile-tmpl.md (notably Decision Profile, Social Proof to Deploy, and Authority Signals to Demonstrate). This can break consistency for downstream workflow/template usage.
Proposed alignment patch
- ### Buyer Overview
+ ### Decision Profile
- **Name:** {name}
- **Role:** {title} at {company}
- **Decision Role:** {decision-maker | influencer | gatekeeper}
- **Personality Type:** {analytical | expressive | driver | amiable}
+ - **Communication Preference:** {data | stories | results | relationships}
...
- ### Approach Recommendations
+ ### Social Proof to Deploy
+ - {specific references, case studies, testimonials}
+
+ ### Authority Signals to Demonstrate
+ - {credentials}
+ - {expertise signals}
+
+ ### Recommended Approach Sequence
- **Opening:** {how to open based on top influence levers}
- **Building Trust:** {specific trust signals to deploy}
- **Creating Urgency:** {scarcity/loss aversion if applicable}
- - **Social Proof to Deploy:** {specific references, case studies, testimonials}🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@squads/negotiation/tasks/profile-buyer.md` around lines 80 - 110, The
output_format block in this file diverges from the canonical buyer-profile-tmpl
schema; update the sections to exactly match buyer-profile-tmpl by replacing or
renaming the current headers and fields (e.g., change "Decision Role" to
"Decision Profile", add "Authority Signals to Demonstrate", and ensure "Social
Proof to Deploy" matches the template wording and order), preserve the existing
placeholder tokens (e.g., {name}, {title}, {company}, {top principle}) and table
structure but align column/section names and section ordering with
buyer-profile-tmpl so downstream consumers see the identical contract.
Comment on lines
+41
to
+44
| ### Timeline | ||
| | Phase | Duration | Deliverable | | ||
| |-------|----------|-------------| | ||
| | {Phase 1} | {time} | {output} | |
There was a problem hiding this comment.
Add a blank line before the Timeline table to satisfy markdownlint.
At Line 42, the table starts immediately after the heading; lint rule MD058 expects a blank line around tables.
Proposed fix
### Timeline
+
| Phase | Duration | Deliverable |
|-------|----------|-------------|
| {Phase 1} | {time} | {output} |📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
Suggested change
| ### Timeline | |
| | Phase | Duration | Deliverable | | |
| |-------|----------|-------------| | |
| | {Phase 1} | {time} | {output} | | |
| ### Timeline | |
| | Phase | Duration | Deliverable | | |
| |-------|----------|-------------| | |
| | {Phase 1} | {time} | {output} | |
🧰 Tools
🪛 markdownlint-cli2 (0.21.0)
[warning] 42-42: Tables should be surrounded by blank lines
(MD058, blanks-around-tables)
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@squads/negotiation/templates/proposal-tmpl.md` around lines 41 - 44, The
Timeline table starts immediately after the "### Timeline" heading which
violates MD058; fix by inserting a single blank line between the "### Timeline"
heading and the table (the table beginning with "| Phase | Duration |
Deliverable |") so the heading and the table are separated by an empty line.
Comment on lines
+31
to
+39
| phase_3: | ||
| name: "Tactical Playbook" | ||
| executor: chris-voss | ||
| task: "tasks/negotiate-deal.md" | ||
| elicit: true | ||
| input_from: ["phase_1", "phase_2"] | ||
| output: "Accusation audit, calibrated Qs, labels, Ackerman plan" | ||
| checkpoint: | ||
| gate: "Complete playbook with scripts" |
There was a problem hiding this comment.
Phase 3 input contract is under-specified for tasks/negotiate-deal.md.
Line 36 only passes outputs from phase_1 and phase_2, but the negotiation task requires additional inputs (Deal context, Current terms / proposal, Known objections). This can leave phase_3 without required structured context.
💡 Possible workflow fix
phases:
+ phase_0:
+ name: "Deal Context Intake"
+ executor: negotiation-chief
+ task: "tasks/define-offer.md"
+ elicit: true
+ output: "Deal context, current terms/proposal, known objections"
+ checkpoint:
+ gate: "Deal context and proposal are documented"
+
phase_1:
name: "Buyer DNA Analysis"
@@
phase_3:
name: "Tactical Playbook"
executor: chris-voss
task: "tasks/negotiate-deal.md"
elicit: true
- input_from: ["phase_1", "phase_2"]
+ input_from: ["phase_0", "phase_1", "phase_2"]🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@squads/negotiation/workflows/wf-negotiate-deal.yaml` around lines 31 - 39,
phase_3 currently only lists input_from: ["phase_1", "phase_2"] but the task
tasks/negotiate-deal.md needs explicit structured inputs (Deal context, Current
terms / proposal, Known objections); update phase_3 to supply those required
fields by either expanding input_from to reference the specific producers (e.g.,
["phase_1.deal_context", "phase_2.current_terms", "phase_2.known_objections"])
or add an inputs section that maps Deal context, Current terms / proposal, and
Known objections to the appropriate upstream outputs so tasks/negotiate-deal.md
always receives the required structured context.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
negotiation— Full sales negotiation pipeline with 9 agents, 10 tasks, 2 workflows, 3 templates, and 2 checklistsThe Expert Team
negotiation-chiefrobert-cialdinichet-holmesjeb-blountoren-klaffneil-rackhamchris-vossjim-campwilliam-uryWhat's Included
wf-full-pipeline(end-to-end 8 phases) andwf-negotiate-deal(focused negotiation)Test plan
/negotiation:negotiation-chief/negotiation:chris-voss)*full-pipelineworkflow end-to-end*negotiate-dealworkflowsquad.yamlandconfig.yamlschemas🤖 Generated with Claude Code
Summary by CodeRabbit
Release Notes