GentBajko/capstone / skills/core
core
VerifiedInternal to the capstone suite - carries the shared rules (references/) and scripts every capstone subcommand reads; it exists so npx-skills installs ship them alongside the command skills. Not meant to be invoked directly; when invoked anyway, run references/../scripts/help.sh and output its stdout verbatim.
references/style.md
View on GitHub# Writing Rules for Generated Docs
The audience is a future Claude session with zero context that must answer
architecture questions without exploring the repo.
## Voice
- Declarative facts only. "Commands are validated in `server/ws.py:88`",
never "commands should be validated" or "consider validating".
- No recommendations, no judgments, no TODOs, no hedging ("probably",
"seems to", "appears").
- If something could not be determined, state that explicitly:
"Retry behavior: not implemented anywhere in `src/net/`."
## Density
- Every claim carries a pointer: `path/to/file.py:123` or `path/to/dir/`.
- Prefer a table for enumerable facts (models, endpoints, dependencies);
prose only for relationships a table cannot show.
- No filler: delete any sentence that does not help the reader locate or
understand code. No introductions, no summaries of summaries.
- The index (`<index_file>`) stays under ~2 pages. Topic files have no
hard cap but must earn their length: if a section repeats the index,
cut it.
- `changelog.md` is exempt from every length rule: it is never
condensed, summarized, or pruned.
## Naming
- Use the codebase's own names verbatim (`GenerateMapCommand`, not "the
map generation command object").
- First mention of a symbol gives its definition site; later mentions do
not repeat it.
## Structure
- Follow the required section templates in `topics.md` exactly: future
sessions rely on stable headings to jump to what they need.
- Absent things are facts too: record them where a reader would look
("No message queue; all communication is synchronous websocket frames").
- Label shallow coverage: when an area was only skimmed, say so
("shallow: static assets, not inventoried") so readers can distinguish
small-and-boring from not-examined.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |