ScopeCairn
An open-source code scoping tool for AI coding agents. Through commands such as context, impact, read and scan, it provides structured code maps, impact analysis and symbol-level reading, helping agents precisely locate the files and symbols they need.
Timeline (last 90 days)
ScopeCairn is an open-source tool designed around the philosophy of "understand first, then explore", helping AI coding agents precisely locate the scope of code changes through a three-step loop.
ScopeCairn's context command returns a structured map, including relevant files (graded HIGH/MEDIUM), key symbols and scoring rationale, dependency relationships, relevant tests, circular reference warnings, Scope grading (required/optional/protected), and Task Guard.
ScopeCairn's context command supports a --knowledge knowledge mode for explanation, summarization, and audit tasks; this mode does not output Scope and Guard coding rules, and only gives a summary, reasons, and usage.
ScopeCairn's impact command outputs four dimensions: direct callers, indirect impact (2 hops), affected tests, and risk level (HIGH/MED/LOW).
ScopeCairn specifies a hard rule: if fan-in (number of times called) is ≥ 10, or the symbol is marked as "protected", stop and confirm with a human.
ScopeCairn's read symbol command returns an excerpt of about 40 lines (including the docstring) rather than the complete file, to reduce token consumption.
ScopeCairn's path command returns the shortest call chain between two points, helping developers identify the path that actually triggers the problem among multiple possible entry points.
ScopeCairn's history command can retrieve similar historical changes with diffs.
After use, ScopeCairn must run the scan command to rebuild the index, otherwise the next context query will be based on outdated data.
ScopeCairn does not use embeddings, does not connect to the network, and does not report telemetry; it relies on static analysis and graph algorithms (such as PageRank popularity ranking, cycle detection, and community detection).
2 more timeline events
All Facts (12)
ScopeCairn is an open-source tool designed around the philosophy of "understand first, then explore", helping AI coding agents precisely locate the scope of code changes through a three-step loop.
50%UnverifiedScopeCairn's context command returns a structured map, including relevant files (graded HIGH/MEDIUM), key symbols and scoring rationale, dependency relationships, relevant tests, circular reference warnings, Scope grading (required/optional/protected), and Task Guard.
50%UnverifiedScopeCairn's context command supports a --knowledge knowledge mode for explanation, summarization, and audit tasks; this mode does not output Scope and Guard coding rules, and only gives a summary, reasons, and usage.
50%UnverifiedScopeCairn's impact command outputs four dimensions: direct callers, indirect impact (2 hops), affected tests, and risk level (HIGH/MED/LOW).
50%UnverifiedScopeCairn specifies a hard rule: if fan-in (number of times called) is ≥ 10, or the symbol is marked as "protected", stop and confirm with a human.
50%UnverifiedScopeCairn's read symbol command returns an excerpt of about 40 lines (including the docstring) rather than the complete file, to reduce token consumption.
50%UnverifiedScopeCairn's path command returns the shortest call chain between two points, helping developers identify the path that actually triggers the problem among multiple possible entry points.
50%UnverifiedScopeCairn's history command can retrieve similar historical changes with diffs.
50%UnverifiedAfter use, ScopeCairn must run the scan command to rebuild the index, otherwise the next context query will be based on outdated data.
50%UnverifiedScopeCairn does not use embeddings, does not connect to the network, and does not report telemetry; it relies on static analysis and graph algorithms (such as PageRank popularity ranking, cycle detection, and community detection).
50%UnverifiedScopeCairn is installed via npm install -g scopecairn and requires Node ≥ 20.
50%UnverifiedThe ScopeCairn project address is https://github.com/DhaniAAA/ScopeCairn.
50%