Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

SOP: Assumption Challenging · yogsoth-ai | DecimalAI

yogsoth-ai/assumption-challenging

NewSkillScoreMaintained

Challenge each assumption's validity — shared cross-repo SOP

researchby yogsoth-ai

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Research pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/yogsoth-ai-assumption-challenging/SKILL.md

decimalai skills pull yogsoth-ai-assumption-challenging --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/yogsoth-ai-assumption-challenging/SKILL.md https://app.decimal.ai/s/yogsoth-ai-assumption-challenging/SKILL.md

Raw file: /s/yogsoth-ai-assumption-challenging/SKILL.md ↗ · machine-readable summary at /s/yogsoth-ai-assumption-challenging.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

23 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 7d ago

No — -9% lift+15% tokens0% turns4% of whole cases pass with skill (1/23)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-04	✓→✗	▼ Worse	40%	0%
case-05	✓→✗	▼ Worse	85%	0%
case-03	✓→✓	＝ Same ✓	21%	0%
case-01	✗→✗	＝ Same ✗	20%	0%
case-02	✗→✗	＝ Same ✗	111%	0%

Showing 5 of 23 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

yogsoth-ai / de-anthropocentric-research-engine

skills/assumption-challenging/SKILL.md

View on GitHub ↗

License Apache-2.0Imported 8/3/2026Version v1

Skill contents

SKILL.md

SOP: Assumption Challenging

Challenge each assumption's validity using systematic questioning. Determine which assumptions are well-founded, which are fragile, and which are likely false.

Subagent — spawned via subagent-spawning/spawn-agent skill.

Shared: This SOP is used across multiple strategies (assumption-constraint, conflict-resolution, constraint-breaking) and campaigns.

Available SOPs

Optional, no fixed order; the final leaf is always a sop.

| SOP | When to use | | --- | --- | | spawn-agent | Spawn a customized CC subagent with full MCP tool access. Used by SOPs that declare execution: subagent. |

Want to see if yogsoth-ai/assumption-challenging would activate on your phrasing? Try it before you install — no LLM call, no signup.

Latest benchmark · publisher runVerified

8/22/2026, 06:49 UTC · 23 cases

Lift

-9%

4% vs 13%

Agent turns

0%

1.0 vs 1.0

Tokens

+15%

2,750 vs 2,395.83

pass → pass · 1fail → fail · 20pass → fail · 2

Per-case results (23)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-01 I am putting together a product roadmap for our new fintech app, but I'm worried we're relying on unexamined premises about user behavior and regulatory timelines. Could you analyze our underlying beliefs about the market launch, test their strength through structured interrogation, and categorize them based on how solid or risky they are?	fail→fail	18,285	21,099	+15%	1	1	0%	2,908	3,496	+20%	0	0	—
case-02 Our marketing campaign strategy relies on several implicit premises regarding customer acquisition channels and conversion rates. Please conduct a thorough review of these baseline assumptions by scrutinizing each one, identifying which hold up under scrutiny, which are on shaky ground, and which appear untruthful.	fail→fail	8,245	14,693	+78%	1	1	0%	1,239	2,620	+111%	0	0	—
case-03 Draft a formal risk management log for an enterprise cloud migration project, listing risk IDs, risk descriptions, likelihood scores, impact scores, and mitigation owners.	pass→pass	12,465	14,757	+18%	1	1	0%	2,175	2,629	+21%	0	0	—
case-04 We are writing a project charter for a supply chain software upgrade. Generate a list of baseline project assumptions regarding budget availability, steering committee approval, and vendor availability for inclusion in section 4 of the charter.	pass→fail	9,885	13,015	+32%	1	1	0%	1,680	2,357	+40%	0	0	—
case-10 A healthtech startup assumes HIPAA compliance can be achieved in two weeks by adopting a pre-configured cloud template. Perform systematic interrogation on this operational assumption and output the final verdict.	fail→fail	14,871	18,733	+26%	1	1	0%	2,360	3,274	+39%	0	0	—
case-05 Conduct a standard SWOT analysis (Strengths, Weaknesses, Opportunities, Threats) for a regional craft brewery expanding into nationwide e-commerce distribution.	pass→fail	14,557	24,099	+66%	1	1	0%	2,195	4,055	+85%	0	0	—
case-06 Our SaaS company assumes enterprise customers will migrate to our v2 API within 30 days of release and that legacy endpoint support can be deprecated immediately. Scrutinize these operational premises through systematic interrogation, forcing a decision on whether each assumption should be relied upon.	fail→fail	15,348	14,279	-7%	1	1	0%	2,439	2,459	+1%	0	0	—
case-07 We are planning a hiring expansion for our cybersecurity team assuming top-tier security analysts will accept 100% in-office requirements in a mid-sized city. Challenge this hiring strategy premise with targeted questioning and categorize its reliability.	fail→fail	12,695	14,991	+18%	1	1	0%	1,949	2,388	+23%	0	0	—
case-08 An e-commerce retailer assumes that switching from freight to air cargo will cut delivery times by 80% while costs only rise by 10%. We need you to challenge these two cost and timeline assumptions using systematic questioning and assess their validity.	fail→fail	16,727	17,770	+6%	1	1	0%	2,613	3,080	+18%	0	0	—
case-09 Our team believes that enterprise procurement cycles will shrink from 90 days to 14 days simply because we added self-serve credit card checkout. Analyze this assumption by challenging its underlying premises.	fail→fail	16,412	13,604	-17%	1	1	0%	2,663	2,311	-13%	0	0	—
case-11 Our retail expansion strategy assumes customer foot traffic in physical malls will increase by 25% year-over-year due to our new loyalty program. Conduct an assumption challenge on this premise.	fail→fail	15,983	16,494	+3%	1	1	0%	2,376	2,767	+16%	0	0	—
case-12 A renewable energy firm assumes battery storage hardware costs will drop 50% in the next quarter despite raw material shortages. Interrogate this supply chain assumption systematically and assign its validity state.	fail→fail	16,378	11,007	-33%	1	1	0%	2,562	1,776	-31%	0	0	—
case-13 Our game studio assumes players will spend an average of $15 monthly on cosmetic microtransactions in a single-player offline title. Scrutinize this monetization premise through systematic questioning.	fail→fail	18,411	14,912	-19%	1	1	0%	2,837	2,461	-13%	0	0	—
case-14 A logistics platform assumes truck drivers will voluntarily adopt a new mobile tracking app that runs continuously in the background without cellular data compensation. Challenge this operational adoption assumption.	fail→fail	15,555	17,785	+14%	1	1	0%	2,209	2,813	+27%	0	0	—
case-15 Our mobile app team assumes iOS users will accept push notification permissions at a 90% rate if presented with custom graphics upon first launch. Conduct systematic questioning on this UX assumption.	fail→fail	14,366	18,507	+29%	1	1	0%	2,290	3,171	+38%	0	0	—
case-16 A commercial real estate developer assumes office space leasing rates in central downtown will return to 2019 peak levels by Q3. Interrogate this market forecast assumption and categorize its validity.	fail→fail	15,779	13,727	-13%	1	1	0%	2,439	2,311	-5%	0	0	—
case-17 Our hardware startup assumes factory yield rates will reach 99% on the first production run of a complex wearable device. Interrogate this manufacturing premise.	fail→fail	22,192	14,537	-34%	1	1	0%	3,119	2,423	-22%	0	0	—
case-18 A financial trading firm assumes zero-latency execution can be achieved over standard public internet endpoints across international regions. Challenge this latency assumption systematically.	fail→fail	17,792	16,802	-6%	1	1	0%	2,802	2,717	-3%	0	0	—
case-19 Our non-profit assumes donor retention will increase 40% solely by redesigning our quarterly email newsletter layout. Scrutinize this engagement belief.	fail→fail	15,246	13,703	-10%	1	1	0%	2,340	2,221	-5%	0	0	—
case-20 An AI company assumes open-source model weights will remain completely unregulated across all foreign jurisdictions for the next five years. Conduct systematic questioning on this regulatory belief.	fail→fail	16,266	18,167	+12%	1	1	0%	2,524	2,916	+16%	0	0	—
case-21 Our streaming video platform assumes users will tolerate 3 minutes of unskippable ads every 5 minutes on a premium paid tier. Interrogate this monetization assumption.	fail→fail	17,848	16,658	-7%	1	1	0%	2,754	2,831	+3%	0	0	—
case-22 An automotive manufacturer assumes consumers will pay a $500 monthly subscription fee for basic heated seat hardware already installed in the vehicle. Challenge this product strategy belief.	fail→fail	13,896	20,919	+51%	1	1	0%	2,052	3,328	+62%	0	0	—
case-23 Our enterprise IT group assumes employees will willingly switch from Slack to a custom internal chat app with no third-party integrations. Conduct systematic interrogation on this user adoption premise.	fail→fail	17,660	18,352	+4%	1	1	0%	2,579	2,846	+10%	0	0	—

These cases were also screened by the skill’s intent / safety pass.See the Trust review ↗Reproduce this benchmark in Colab ↗

How this skill was measured

DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted. The headline lift of -9 percentage points is the difference between those two pass rates over the 23 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 13%
Pass rate with the skill: 4.4%
Cases attempted: 23
Cases counted toward lift: 23
Cases that regressed: 2
Skill version measured: v1
Run completed: 8/22/2026, 06:49 UTC

SkillScore blends 4 signalsno signals yet

BenchmarkRun a verified benchmark with positive lift

Live evalNeeds ≥20 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

Efficiency · separate axis (at equal correctness)+0% turns+21% tokens

Shown for transparency; not credited unless a verified run shows no correctness regression.

No signals yet — run a benchmark or collect evaluated traces to earn a score.

Verified on gemini-3.6-flash · re-tested 7 days ago

Other measured skills in the registry, with their headline benchmark lift.