senior-project-intelligence
Skill2 starsChanged 15 days agoApache-2.0View on GitHub
Use when starting substantive work in an unfamiliar repository, after a material architecture or stack change, or when sessions keep rediscovering the same project facts. Not for small local edits that can be done from the files at hand.
In this repository7 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- senior-code-reviewSkill
- senior-debuggingSkill
- senior-engineeringSkill
- senior-finishSkill
- senior-skill-authoringSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 13
If available, run senior inspect for deterministic stack facts.
- Distinctive: line 14
Read only the entry points needed to understand architecture boundaries.
- Very specific: line 16
Run senior init to create .senior/ when it does not exist yet.
- Distinctive: line 45
Before finishing, confirm the stored context is accurate against the current repository, concise, free of secrets, and useful across …
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
Project Intelligence
Core rule: store only facts that will save a future session from a mistake.
Process
- Read repository-level instructions.
- If available, run
senior inspectfor deterministic stack facts. - Read only the entry points needed to understand architecture boundaries.
- Separate durable facts from transient implementation details.
- Run
senior initto create.senior/when it does not exist yet. - Record a decision in
.senior/decisions.mdonly when it constrains future work, dated and in one or two lines.
Good durable facts
- language, framework, package manager;
- database/ORM and auth provider;
- deployment target and service boundaries;
- the canonical verification commands;
- conventions that are easy to violate;
- decisions such as "use Drizzle; do not introduce Prisma".
Do not store
- secrets, credentials or personal data;
- raw conversation transcripts;
- temporary debugging hypotheses;
- large file inventories;
- facts that are obvious from one local file, unless they keep causing mistakes;
- volatile data that
senior inspectcan re-detect.
| If you notice yourself thinking… | Do this instead |
| "Let me map the whole repository first" for a local change | Read the files the task touches |
| "This might be useful someday" | Store it only if a future mistake depends on it |
| "I'll paste the env file so the next session has it" | Never; reference where configuration lives instead |
Verification
Before finishing, confirm the stored context is accurate against the current repository, concise, free of secrets, and useful across sessions.
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.