plugins/negotiation/skills/reading-people/SKILL.md
Use when user wants to read body language, nonverbal cues, comfort/discomfort signals during in-person/video meeting, pitch, interview, or negotiation. Joe Navarro methodology — limbic system, baseline, feet→face honesty hierarchy. Trigger phrases — "číst lidi", "řeč těla", "body language", "nonverbalni komunikace", "pozorovat", "Navarro", "co znamená když...". Do NOT use for verbal tactics (use tactical-empathy), strategic prep (use batna-strategy), or written communication (use written-negotiation).
npx skillsauth add petrogurcak/skills reading-peopleInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
This skill teaches Joe Navarro's methodology for reading people during business negotiations. It is built around one binary question:
Am I seeing comfort or discomfort?
That is the entire frame. Not "are they lying?", not "are they angry?", not "what emotion is this?" — only comfort vs. discomfort. Navarro himself made this pivot explicit in The Dictionary of Body Language (2018), in a chapter titled "Detecting Deception: Proceed with Caution!":
"There are no nonverbal behaviors that, in and of themselves, are clearly indicative of deception. As my friend and researcher Dr. Mark G. Frank repeatedly has told me, 'Joe, unfortunately, there is no Pinocchio effect when it comes to deception.' With that I must humbly concur. Therefore, in order to sort fact from fiction, our only realistic recourse is to rely on those behaviors indicative of comfort/discomfort, synchrony, and emphasis to guide us."
— Joe Navarro, The Dictionary of Body Language, 2018
This is not a softening — it is a correction. Navarro spent 25 years in FBI behavioral analysis and concluded that even the best human lie detectors hover near 54% accuracy, which is close to chance (Vrij meta-analyses; DePaulo & Bond, 2006). Microexpressions tested in PMC6158306 (Frontiers in Psychology, 2018) underperformed below 56%. Single-tell signals like gaze aversion are junk — practiced liars often over-maintain eye contact rather than break it.
So this skill is not a lie detector. It is a tool for spotting psychological discomfort — moments where a topic, number, clause, or person makes the counterpart's limbic system flinch. That signal is actionable: it tells you where the friction is, which question to ask next, which clause to revisit. The cause of the discomfort (real concern, unresolved objection, time pressure, room temperature) you discover by asking calibrated questions verbally — not by guessing from body language alone.
The methodology rests on three foundations:
Everything below operationalizes those three foundations for a Czech business setting (investor pitch, klient meeting, supplier negotiation, agency first call, hiring conversation, salary review).
Use this skill when the user is preparing for, or in the middle of, a live face-to-face or video meeting where they need to read the other side. Concretely:
emotional-conflict for the verbal layer)tactical-empathy (Voss methodology). This skill reads bodies; that one feeds words.batna-strategy. You cannot out-read a counterpart whose alternatives are stronger than yours; structure beats observation.written-negotiation. There is no body to read in a Slack thread.Most of the work happens before the counterpart enters the room. Environment design beats real-time observation. Default checklist:
For video calls (Zoom, Teams, Google Meet):
You cannot read someone else's limbic system if your own is in fight/flight. Before the counterpart enters:
If you arrive stressed (rushed, traffic, prior call ran long), spend the first 60 seconds in the bathroom or hallway resetting. Do not enter the room with elevated cortisol — your reads will be wrong because your nervous system is biasing perception toward threat.
The first 3-5 minutes of any meeting are not "small talk." They are data collection. Navarro:
"Establishing a person's baseline behavior is critical because it allows you to determine when he or she deviates from it, which can be very important and informative. Not getting a baseline puts you in the same position as parents who never look down their child's throat until the youngster gets sick. By examining what's normal, we begin to recognize and identify what's abnormal."
— What Every Body Is Saying, 2008
While you talk weather, traffic, and coffee, you silently log:
| Channel | What to log | | ------------------------ | --------------------------------------------------------------------------- | | Blink rate | Slow and even? Fast and fluttery? Estimate baseline blinks per 30 seconds | | Hand position | Visible on table? Hidden in lap? Steepled? Interlaced? Palm-up or palm-down | | Foot direction | Pointed at you? At the door? Locked under chair? Bouncing? | | Voice tempo | Words per minute, pause length, breath depth | | Posture | Leaning forward / back / neutral. Shoulders up or down. Head tilt | | Fidget patterns | None? Pen click? Ring twist? Chair swivel? | | Eye-contact duration | Long stretches? Short glances? Looking-away pattern (up/down/side?) |
Safe baseline topics, ranked by reliability:
Avoid as baseline topics: WWII, communism, Russia/Ukraine, Roma, religion, salaries, family income, healthcare politics. These contaminate the baseline because they introduce loaded emotional content before you have data on the counterpart's neutral state.
A person who fidgets constantly: their stillness is the tell. A person who is naturally still: their fidget is the tell. Without baseline you see fidget and conclude "stressed" — you may have just observed their normal Tuesday morning.
This matters double in CZ business culture, where the cultural baseline is lower amplitude. Czechs deliberately suppress facial and bodily expression — exuberant gesture reads as overselling, suspect, "fake American." So the absolute amplitude bar is lower across the board, and you must read delta from this person's individual lower baseline. Detection sensitivity matters more than expecting a big tell. A subtle lip compression, a half-second hand to neck, a brief torso pull-back — in CZ context these are the same magnitude signal as a US counterpart wiping a sweaty brow.
You memorize it. Five minutes in, you should be able to silently say to yourself:
That is now the reference. From minute 5 onward, you watch only deviations. If Pavel's hands suddenly go under the table when you mention price, that's signal. If his foot stops bouncing the second you state your equity ask, that's signal. Without baseline you would have missed both.
A meeting moves through four phases. Each phase has its own watch list and tactic.
Watch:
Tactic:
Watch:
Tactic:
tactical-empathy skill for the full verbal layer).Watch:
Tactic:
Watch — commitment cluster (genuine yes):
Watch — fake yes / ventral denial cluster:
Tactic for fake yes:
Do not accept the verbal commitment at face value. Slow the close. Ask a calibrated question: "Co vás na tom ještě trápí?" / "Co bychom měli ještě probrat, aby vám to sedělo?" Watch what happens when you give them permission to surface the concern. Often the real objection emerges in the next 60 seconds.
Doorjamb moment. Navarro's Three Minutes to Doomsday documents the FBI tactic of saving the hardest question for the moment the counterpart thinks the meeting is over — handshake done, jacket on, walking to the door. Their guard drops because they think it's over. "Aha, jen ještě jedna věc — kdo na vaší straně rozhoduje o tomhle finálně?" — asked while standing at the door, gets a different answer than asked while seated at the table.
Most CZ business meetings have 2-4 people on the other side. Read them in a deliberate order:
Sometimes the counterpart is so composed (high-trained negotiator, M&A lawyer, senior politician) that visible cues are minimal. Three options:
If still nothing, accept that this person is not readable for you today and rely more on verbal calibrated questions and structural prep (tactical-empathy, batna-strategy). Reading is one channel among several.
For full coverage see ../references/comfort-discomfort-taxonomy.md (master reference, 130+ catalogued behaviors).
The 8-12 cues you will use in 80% of negotiations:
| Cue | What it looks like | What it usually means | | --------------------------------------- | ------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------- | | Suprasternal notch touch | Hand to the dimple between collarbones | Strongest pacifier — distress, threat, insecurity. Especially reliable in women | | Leg cleanser | Palm slides down thigh under table (look for upper-arm/shoulder movement) | High-grade discomfort. "Almost always when confronted with damning evidence" — Navarro | | Lip compression / disappearing lips | Lips narrow, blood drains | True negative sentiment. "It rarely, if ever, has a positive connotation" — Navarro | | Ventral denial | Torso turns / angles away from you | Subconscious distancing — they don't like you, the topic, or what was just said | | Ventral fronting | Torso opens toward you, often with lean forward | Comfort, interest, agreement | | Eye-blocking | Slow blinks, hand to eye, eyebrow rub, squint | Concern, dislike, disagreement, perceived threat — "don't want to see/hear this" | | Foot direction | Where toes point | Where the person wants to go (exit, you, away from a third party) | | Steepling | Fingertips touched in peaked roof, fingers not interlaced | Highest confidence — they are committed to this position. Don't push it; pivot | | Palm-up plea | Palms upward during declaration | Asking to be believed. Treat declarations as soft, not authoritative | | Palm-down emphatic | Palms pressed down on table during statement | Strong commitment, firm position — believed by speaker | | Freeze (turtle effect) | Sudden stillness, shallow breath, shoulders rise toward ears | First-stage limbic stress response — easily missed because it's the absence of motion | | Hooding | Hands interlaced behind head, elbows splayed | Territorial dominance display — claiming alpha in the room |
Cluster rule: a single tell is unreliable. Two or three cues clustering on the same trigger or in sequence increases confidence. Navarro's Commandment 6: "Always try to watch people for multiple tells — behaviors that occur in clusters or in succession."
❌ The lie-detector trap. Reading body language to "catch lies." Navarro himself, after 25 years at FBI, says explicitly: "Even the best experts, including myself, are only a blink away from chance, and have a fifty-fifty probability of being right or wrong. Plainly put, that's just not good enough!" Do not play polygraph. Read for discomfort, ask why with verbal questions.
❌ Single-tell fallacy. "He touched his nose, he's lying." No. He touched his nose because his nose itched, the room is dry, he has hay fever, he just thought of something stressful, or he is processing what you said. Single tells are noise. Wait for clusters.
❌ Ignoring baseline. Comparing the counterpart to yourself, to a "confident person template," or to a generic body-language YouTube video. The only valid comparison is this person now vs. this person five minutes ago. Without baseline, you cannot read.
❌ Staring at the face. The face is the most rehearsed, most controlled part of the body. By the time the face shows it, the feet showed it two seconds earlier and were truer. Start observation low (feet, legs, foot direction) and work up.
❌ Weaponized observation. Calling out cues mid-meeting ("Aha, vy jste si sáhl na krk, takže s tím nesouhlasíte!"). This is not body-language reading — this is being a jerk. The moment the counterpart knows you are watching for tells, they freeze and the data dries up. Navarro's Commandment 10: "When observing others, be subtle about it... your ideal goal is to observe others without their knowing it, in other words, unobtrusively."
❌ Pseudo-scientific certainty. "Crossed arms = closed off." "Looking up-left = lying." These are pop-psych myths, not Navarro. Ekman's microexpression methodology has been heavily criticized (PMC6158306) — accuracy below 56%, frequency too low to be useful. NLP "eye accessing cues" has zero replicated evidence. Avoid this whole register; it makes you look amateur and gets you wrong reads.
❌ Treating one read as enough. "V první minutě si sáhl na krk, takže nesouhlasí s cenou." Maybe. Or maybe his collar itched, or the air conditioning is too cold, or he got bad news on the way over. Wait for the second hit. Two cues on the same trigger = signal. One cue = noise.
❌ Weaponizing this skill against intimate partners, family, or vulnerable people. Reading discomfort in a business setting where both parties are negotiating is fine. Using this as surveillance on a spouse, child, or therapy patient is a violation. Skill is for adversarial-cooperative business contexts, not personal relationships.
❌ Confusing comfort with agreement. A counterpart who is comfortable in your presence is not the same as a counterpart who agrees with your position. They can be perfectly relaxed and still walk away. Comfort = "I trust you in the room." Agreement = "I will do the deal." Different signals, different evidence.
❌ Discounting the "boring meeting" tells. Meetings that feel uneventful often have the highest-information bodies — because no one is performing. The micro-tells in calm meetings are more reliable than the dramatic ones in heated ones (which are confounded by general arousal). Do not dismiss data just because the meeting felt "fine."
This skill helps you observe psychological discomfort during business negotiations so you can ask better verbal questions. That is its only legitimate use.
It is not a tool for:
If the user asks for any of the above, refuse the framing and redirect to a legitimate professional (HR, lawyer, therapist, due-diligence firm).
For full delta tables see ../references/cz-business-culture-deltas.md.
Six adjustments specific to reading bodies in CZ business culture:
Lower amplitude baseline. Czechs deliberately suppress facial and bodily expression. American/Italian-style expressivity reads as oversold and untrustworthy. Therefore: do not expect a big tell. The CZ equivalent of an American shoulder shrug + eye roll is a quarter-second lip compression + half-blink. Sensitivity beats expecting amplitude.
Vykání as baseline. During the first meeting, vykání is the social default. A counterpart who tries to switch to tykání before establishing relationship trust is either testing dominance or genuinely friendly — watch their body for which. A premature tykání without invitation triggers limbic discomfort that you will read as "something off" — it is, but the cause is the social transgression, not deception.
"To bude těžké" body confirmation. When a Czech says "To bude těžké" / "Uvidíme" / "To si musíme rozmyslet" — these almost always mean NO, not maybe. Confirm by reading the body that delivers the line: lip compression, brief eye-block, ventral lean-back, palm-up plea — all consistent with polite refusal. If the body matches the verbal hedge, treat it as no and adjust accordingly. Do not push a "yes" out of "to bude těžké."
Handshake as data. CZ handshake norm is firmer than US. A weak grip from a counterpart who is otherwise composed is a tell — typically discomfort or low confidence on this specific deal. Wet/cold/clammy palm is sympathetic activation, but cold weather causes false positives — discount the signal in winter or after they've been outside.
Hierarchy reads. Czech business decisions roll up to seniority. The person at the table may not have closure authority. Watch who the speaker glances at when a hard question is asked — that's the actual decision-maker, even if they're silent. Watch for the silent person's body: pacifiers, lip compression, head shake — the silent senior is often where the real read is.
Pivo / káva vs. lunch / dinner. Coffee and beer meetings are business; lunch and dinner meetings are relationship. The body baseline shifts radically across these contexts. A counterpart who is steepled-and-firm at the conference table may be ventral-fronted-and-relaxed at the bar two hours later — that is not deception, that is context. Do not generalize lunch behavior to next week's contract review.
For deeper CZ-context training, three sources are canonical:
Tito autoři kalibrují generic Navarro methodology na CZ kontext (vykání, hierarchie, "to bude těžké" idiomy, paper-trust). Pokud máš nějakého z nich v knihovně, čti vedle této skillu — vzájemně se doplňují.
| CZ phrase | Translation | Body confirmation | | ------------------------ | ------------------------------- | -------------------------------------------------- | | "To bude těžké" | NO (polite) | Lip compression + ventral lean-back + soft palm-up | | "Uvidíme" | NO or stalling | Eye-block + foot toward door + brief freeze | | "To si musíme rozmyslet" | NO or deferred decision | Hidden hands + ventral denial + glance to senior | | "Není to standardní" | NO (org pushback) | Steepled + firm voice — they mean it | | "Tykejme si" | Relationship milestone — accept | Eyebrow flash + ventral fronting + open palms | | "Děkujeme za nabídku" | Could be either | Read body — fronting = yes, denial = polite no | | "V principu ano, ale..." | NO with face-saving | Pacifier on "ale" — friction is in what comes next |
When verbal hedge + body denial align, treat it as NO and pivot. When verbal hedge + body fronting, push gently for the surface objection — there is room.
When the user invokes this skill with a scenario, return a phase-by-phase observation checklist tailored to their meeting.
User: "Zítra mám pitch s investorem, první meeting, dva v týmu na druhé straně. Co mám sledovat?"
Skill output:
Pre-meeting (do today / before they arrive):
5-min baseline (first 3-5 minutes — small talk only):
Topics: cesta sem, počasí, kávu / vodu, recent neutral news (sports, weather event). Avoid politics, salaries, history.
Silently log per person:
Do this for both counterparts separately. They will have different baselines.
Opening watch list (first 10 minutes after small talk transitions to business):
Probing — what shifts to expect when discussing valuation/equity:
Closing — commitment cluster vs. fake yes:
Genuine commitment to next step (e.g., "Pošleme term sheet") looks like:
Fake yes (verbal commitment but body says no) looks like:
If you see fake yes: slow the close. Ask: "Co vás na tom ještě trápí?" — give them permission to surface the real objection. The next 60 seconds usually surface it.
Doorjamb question: save your highest-leverage question for the handshake at the door. "Aha, jen ještě jedna věc — kdo finálně rozhoduje na vaší straně?" — gets a different answer than the same question asked at minute 15 seated.
User: "Mám zítra první meeting s potenciálním klientem (e-shop, ~50 zaměstnanců). Jdou dva: marketing manažer a CEO. Mám si udělat pitch na ~90 minut. Co sledovat?"
Skill output (compact):
Pre-meeting:
5-min baseline:
Opening (první 10-15 min business talku):
Probing — kdy se ptáš, čeho dosáhli a co je trápí:
Pushback (když ty navrhneš retainer fee):
Closing:
User: "Zítra mám salary review. Chci říct o 15% víc. Co sledovat na šéfovi?"
Skill output (compact):
Pre-meeting:
Baseline (první 2-3 min):
Probing (když mluvíš o svých výsledcích):
Číslo (moment, kdy řekneš "chtěl bych o 15% víc"):
Closing:
Doorjamb: ve dveřích, "Aha, ještě jedna věc — kdo finálně rozhoduje o salary increases? Jen ať si to dobře připravím, kdyby si mě zavolal." — odhalí, jestli šéf má autoritu, nebo to jde výš.
Case 1 — "Investor mlčí celé Q&A, jen občas přikývne. Kde je tell?"
Read: silent senior je nejdůležitější člověk u stolu. Sleduj jeho mikro-pacifiers (krátký dotyk vousů, lip compression během odpovědi jeho juniora, foot direction). Pokud foot pointuje na dveře, jeho zájem klesá s časem — zrychli a zjednodušuj. Pokud foot pointuje na tebe a má ventral fronting, je s tebou v ringu i bez verbálního příspěvku.
Case 2 — "Klient se na meetingu chová super vstřícně, ale po 3 dnech nepíše."
Pravděpodobnost: fake yes detekovaný real-time, ale tys to v meetingu missnul. Watch back na finální handshake — pokud měl wet/cold palm + brief eye-block + ventral lean-back, byl to fake yes. Zavolej, ne email. Telefonem se ozve "To bude těžké" rychleji než písemně.
Case 3 — "Supplier během reviewu klauzule najednou ztichne na 5 sekund."
Freeze response. Klauzule, která to spustila, je friction point. Re-introduce ji za 2-3 otázky paraphrased. Pokud freeze znova → confirmed. Pak calibrated otázka: "Co konkrétně by na téhle klauzuli muselo vypadat jinak, aby vám to fungovalo?"
Case 4 — "Cofounder při diskusi o equity splitu drží ruce pod stolem celou dobu."
Hidden hands generates suspicion, ale v cofounder kontextu to častěji znamená discomfort z konfliktu, ne deception. Reframe: "Vidím, že tě to úplně nesedí. Co tě na tom rozdělení trápí nejvíc?" — explicitně otevřít prostor. Tady přechází k emotional-conflict skillu.
(The skill returns 5-8 specific items per phase. Substitute scenario as needed: klient pitch, supplier negotiation, agency first meeting, salary review, hiring conversation, M&A talks. The phase structure is identical; the trigger topics and decision-makers shift.)
This skill is built on a careful read of what nonverbal-reading research does and does not support. Honesty matters because false confidence in this domain causes real harm — innocent people convicted, deals lost on misread tells, partners falsely accused.
Navarro himself, in The Dictionary of Body Language (2018) and in The End of Detecting Deception (Psychology Today, July 2018; jnforensics.com), explicitly aligns with the critics on deception:
"There are no nonverbal behaviors that, in and of themselves, are clearly indicative of deception... Even those who look for clusters of behaviors would also be wrong — there are no clusters of behaviors indicative of deception... When it comes to detecting deception, even the best experts, including myself, are only a blink away from chance."
What body language can reliably tell you: psychological discomfort — stress, anxiety, concern, tension. The cause of that discomfort you discover by asking calibrated questions, not by guessing from the body. This aligns with Vrij's "cognitive approach" — increase cognitive load via questioning, then watch verbal content for inconsistencies, not body cues.
This skill is framed around reading discomfort to find friction points and unresolved concerns — not catching liars. That position is:
If the user asks "is this person lying to me?" — refuse the framing. Reframe to "where is this person uncomfortable, and what does that tell us about which questions to ask next?"
../references/comfort-discomfort-taxonomy.md — primary lookup table for cues, limbic hierarchy, pacifying behaviors, honesty hierarchy, hand gestures, eye behavior, territorial display. Used during meeting prep and for deeper investigation of specific cues.../references/cz-business-culture-deltas.md — CZ-specific cultural adjustments (handshake, vykání, "to bude těžké" idiom translation, decision-maker hierarchy, meal etiquette, written agreements). Used during pre-meeting prep and CZ-context interpretation.development
Builds a pre-launch social proof strategy through structured beta programs using D'Souza Brain Audit interviews. Use when launching new products/services and need compelling testimonials, planning a beta cohort, designing interview questions to harvest objection-busting social proof, improving video testimonials for landing pages, or designing case studies with metrics. Trigger phrases include "beta tester program for testimonials", "pre-launch social proof", "Brain Audit testimonial framework", "case study harvest", "reverse testimonial", "video testimonial mechanics", "social proof landing page", "sběr referencí", "beta tester program", "testimonial pro landing page", "social proof před launchem", "rozhovor s klientem", "case study sběr", "reference před spuštěním". NOT for ongoing case study production (use growth-hacking case-study approach), offer design (use offer-creation), or conversion optimization (use ux-optimization).
development
Use when planning a product launch and the product type is unclear or could be either generic (SaaS/app/physical) or info-product. Routes between marketing:launch-strategy (generic launches) and marketing:info-product-launch (courses, memberships, ebooks, cohorts, communities). Trigger phrases - "launch", "spuštění", "go-to-market", "product launch", "release strategy", "uvedení na trh", "launch plan", "spuštění produktu", "launch sequence", "launch strategy". Do NOT trigger when product type is already clear (use specific skill directly).
testing
Specialized 8-week launch cadence for info-products — online courses, cohort programs, memberships, communities, ebooks, masterminds. Combines Jeff Walker's Product Launch Formula (Seed/Internal/JV variants, PLC sequence, open-cart day-by-day) with Stu McLaren's membership mechanics (closed cart, Success Path) and Hormozi Grand Slam Offer stacking. Use when planning "launch online kurzu", "info-product launch", "PLF launch", "course launch", "membership launch", "cohort launch", "ebook launch", "open cart close cart", "8-week launch of online course", "beta cohort to launch sequence", "spuštění kurzu", "launch členské sekce", "open cart strategie". Differentiates from marketing:launch-strategy (generic SaaS/app launches) — info-product-specific. NOT for SaaS launches, physical products, or services.
development
Use when releasing an Expo/React Native mobile app to App Store and Google Play - covers eas submit, ASC "Submit for Review", Play promote Internal→Production, OTA update, and decoding common silent failures (Apple agreement expiry, missing English locale, Background Location declaration, web bundle failure on react-native-maps).