cross-file-relationship-synthesis
Skill220 starsChanged 20 days agoMITView on GitHub
Use when several files have been read independently but a conclusion depends on how control or data crosses them: build a typed, directed, line-anchored map of calls, imports, producer-consumer and config edges; identify cut points, blast radius, orphans, and traversal-backed conclusions. It maps arbitrary repository relationships; it does not settle privacy compliance or prescribe graph surgery.
In this repository179 other files it gives its agents.
- AGENTS.mdAGENTS.md
- chrono/CLAUDE.mdCLAUDE.md
- CLAUDE.mdCLAUDE.md
- accessible-media-authoringSkill
- agentic-safety-auditSkill
- agent-prompt-engineeringSkill
- audio-event-map-authoringSkill
- auto-scaffoldSkill
- chain-constructSkill
- chain-construct-smart-contractSkill
- claim-verificationSkill
- code-reachability-auditSkill
- code-review-loopSkill
- color-theorySkill
- conversation-designSkill
- copy-refinementSkill
- cosmos-sdk-audit-checklistSkill
- cross-chain-bridge-auditSkill
- data-flow-traceSkill
- dedup-prior-art-checkSkill
- defi-invariant-checkSkill
- dependency-cycle-auditSkill
- dependency-health-triageSkill
- detection-as-codeSkill
- diff-aware-semgrep-scanSkill
- differential-reviewSkill
- dimensional-analysis-checkSkill
- dual-level-retrievalSkill
- durable-nonce-exploitationSkill
- erc1271-revert-data-checkSkill
- error-based-sstiSkill
- evm-audit-flowSkill
- figma-implement-designSkill
- findings-filterSkill
- forensic-timeline-authoringSkill
- game-design-fundamentalsSkill
- game-mechanics-balancingSkill
- head-tailSkill
- incident-response-runbookSkill
- interactive-audio-designSkill
- interface-ambiguity-checkSkill
- keyword-clusteringSkill
- knowledge-base-integrationSkill
- known-advisory-backport-checkSkill
- layered-analysis-loopSkill
- level-design-patternsSkill
- locale-adaptationSkill
- multi-stance-audit-fanoutSkill
- narrative-structureSkill
- osint-platform-auditSkill
- parser-differential-route-confusionSkill
- persuasive-technical-writingSkill
- platform-complianceSkill
- player-engagement-psychologySkill
- pre-audit-threat-modelSkill
- program-rubric-lookupSkill
- read-only-reentrancy-checkSkill
- requirements-elicitationSkill
- review-severity-ladderSkill
- rule6-rights-gateSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 16
Look specifically for the edges that contradict the obvious structure: the secondary writer, the path that bypasses the abstraction, the …
- Distinctive: line 18
Check the map for orphans and dead ends, and resolve each: either an edge is missing from the map, or the file genuinely is unreachable.
- Distinctive: line 22
Every file in scope has a stated one-line role.
- Distinctive: line 23
Every edge is typed, directed, and anchored at both ends with file and line.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Cross-File Relationship Synthesis
Turn a set of independently-read files into an explicit relationship map, so conclusions rest on traced edges rather than on the impression that the files are related.
Steps
- List the files in scope and give each a one-line role. A file whose role you cannot state in one line has not been understood well enough to relate to others.
- Name the edge kinds you will trace before tracing them — calls, imports, writes-then-reads, schema producer/consumer, config-to-consumer, generator-to-generated. Untyped "related to" edges hide the actual dependency.
- Trace each edge in the direction data or control actually flows, and record the file and line at both ends. An edge without two anchors is a hypothesis.
- Distinguish edges you observed from edges you inferred from naming. Matching names are a lead to verify, never an edge.
- Look specifically for the edges that contradict the obvious structure: the secondary writer, the path that bypasses the abstraction, the consumer nobody registered. These decide correctness more often than the primary path does.
- Identify the cut points — files whose change propagates furthest — and state the blast radius of each as a concrete list of affected files.
- Check the map for orphans and dead ends, and resolve each: either an edge is missing from the map, or the file genuinely is unreachable. Never leave the ambiguity unexplained.
- State the conclusion as a traversal over the map, so a reader can follow the same path and reach the same result.
Acceptance
- Every file in scope has a stated one-line role.
- Every edge is typed, directed, and anchored at both ends with file and line.
- Observed edges and name-inferred edges are visibly separated.
- Cut points are named with an explicit blast-radius file list.
- Orphans and dead ends are each resolved as missing-edge or genuinely-unreachable.
- The final claim is expressed as a walk over recorded edges, not as a summary impression.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

