Kshitijpalsinghtomar/depth-skills

Cognitive architecture for AI agents. 19 skills that force language models past surface-level reasoning into genuine depth. Compatible with Claude Code, Cursor, Gemini CLI, Copilot, and any agent that reads markdown.

4Stars on the repository
26Mods indexed here, across every type
3mo agoLast push, which is what freshness is scored on
MITLicence, which decides whether bodies are shown

api-designer

01

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Designs API contracts for consumer experience first — making correct usage obvious and incorrect usage impossible.

not rated 4 3mo ago A 22 tokens original MIT

copy-engineer

02

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Replaces adjectives with evidence, claims with demonstrations, and promises with proof for conversion-oriented writing.

not rated 4 3mo ago A 23 tokens original MIT

adversary

03

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Forces the model to build a genuine prosecution case against its own answer before delivering it.

not rated 4 3mo ago A 22 tokens original MIT

anchor

04

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Detects and corrects objective drift during extended tasks by periodically checking work against the original request.

not rated 4 3mo ago A 21 tokens original MIT

clarify

05

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Decides whether to ask clarifying questions or proceed with an answer, optimizing for information value vs. delay cost.

not rated 4 3mo ago A 26 tokens original MIT

conductor

06

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Self-selects and sequences the right depth-skills proportional to task consequence, preventing both under- and over-analysis.

not rated 4 3mo ago A 26 tokens original MIT

contradict

07

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Cross-compares every claim in an output to detect internal contradictions that sequential generation hides.

not rated 4 3mo ago A 21 tokens original MIT

deep-think

08

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Forces the model to interrupt its surface response and activate deeper knowledge pathways before answering.

not rated 4 3mo ago A 21 tokens original MIT

descend

09

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Verifies whether the problem was correctly identified before solving, auditing pattern matches against first principles.

not rated 4 3mo ago A 21 tokens original MIT

diverge

10

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Forces generation of genuinely different solution paths before committing, preventing single-path pattern gravity.

not rated 4 3mo ago A 20 tokens original MIT

emergence

11

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Analyzes what components create together that none intended — feedback loops, contention, timing bugs, and assumption collisions.

not rated 4 3mo ago A 26 tokens original MIT

excavate

12

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Digs beneath logic to inspect unchecked premises, rating each assumption by collapse severity.

not rated 4 3mo ago A 19 tokens original MIT

fidelity

13

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Prevents lossy compression from erasing conditions, exceptions, and uncertainties during summarization.

not rated 4 3mo ago A 21 tokens original MIT

invert

14

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Flips constraints and beliefs to expose false walls and test plan robustness against inverted worldviews.

not rated 4 3mo ago A 20 tokens original MIT

negative-space

15

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Illuminates what the answer never addressed — dimensions, stakeholders, and failure categories invisible from the activation path.

not rated 4 3mo ago A 25 tokens original MIT

provenance

16

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Tags every claim as fact, inference, or guess and computes calibrated confidence to prevent epistemic flattening.

not rated 4 3mo ago A 25 tokens original MIT

reframe

17

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Solves the same problem through multiple independent representations, then triangulates structural truth.

not rated 4 3mo ago A 19 tokens original MIT

shallow

18

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Prevents overthinking by matching cognitive investment to consequence - the intentional opposite of DEEP-THINK.

not rated 4 3mo ago A 24 tokens original MIT

teach

19

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Uses the Feynman technique - explain simply to detect where knowledge gaps exist.

not rated 4 3mo ago A 18 tokens original MIT

temporal

20

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Evaluates decisions across multiple plausible futures, analyzing option value and temporal regret before commitment.

not rated 4 3mo ago A 20 tokens original MIT

threshold

21

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Gates commitment on irreversible decisions by measuring reversal cost and enforcing proportional exploration depth.

not rated 4 3mo ago A 18 tokens original MIT

mobile-engineer

22

Kshitijpalsinghtomar/depth-skills

Skill Claude CodeCodex

Re-derives every decision for the mobile context — thumb zones, network reality, attention budgets, and input constraints.

not rated 4 3mo ago A 28 tokens original MIT

At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: