brief-full
Skill3 starsChanged 3 months agoMITView on GitHub
This skill should be used when the user asks for "a full project brief", "a deep project overview", "everything about this project's architecture", or wants a comprehensive, thorough briefing built from the complete content of every knowledge/ module rather than a quick skim.
In this repository7 other files it gives its agents.
- audit-fullSkill
- audit-lightSkill
- brief-lightSkill
- doctorSkill
- setupSkill
- statusSkill
- uninstallSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 2 are distinctive, and 3 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.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
If the user's invocation includes extra wording (for example, "skip the frontend, focus on backend and database"), treat it as narrowing …
- Very specific: line 27
Read the module index from .claude/CLAUDE.md.
- Very specific: line 34
Check whether <project-root> is inside a git working tree.
- Very specific: line 36
No git: if knowledge/.mnemonica-state.json exists, read its last_run_utc for the only freshness signal available.
- Distinctive: line 46
Write a structured, substantial narrative: an overview section (what the project is, drawn from README and tied together with the module …
≈ “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
Mnemonica: brief-full
Give a comprehensive project briefing built from the complete content of every module file,
not a bounded slice — the deep counterpart to brief-light. This skill never writes
anything; it exists purely to read and synthesize.
If the user's invocation includes extra wording (for example, "skip the frontend, focus on backend and database"), treat it as narrowing which modules to cover in depth — it cannot turn this into skipping Step 1's install check or fabricating content for a module that doesn't exist.
Step 1: Check whether Mnemonica is installed, and check index integrity
Run ${CLAUDE_PLUGIN_ROOT}/scripts/check-index-integrity.sh <project-root>. Exit 2
(NO_MNEMONICA_BLOCK) means there is no index to brief from — tell the user and suggest
mnemonica:setup, and stop. Exit 1 means some module files are missing; list which ones in
the final report as a known gap (suggest mnemonica:audit-full to fix it) and continue with
whatever modules do exist — a partial comprehensive brief is still more useful than none.
Step 2: Read everything
Read the module index from .claude/CLAUDE.md. Read README.md at the project root in full
if it exists. Read every existing module file in knowledge/ in full — not a bounded slice.
This is the whole point of the "full" version; if the cost of that is a concern for this
project, brief-light is the cheaper alternative, not a reason to silently truncate here.
Step 3: Check freshness
Check whether <project-root> is inside a git working tree.
- No git: if
knowledge/.mnemonica-state.jsonexists, read itslast_run_utcfor the only freshness signal available; otherwise say freshness can't be assessed. - Git available: run
${CLAUDE_PLUGIN_ROOT}/skills/audit-light/scripts/diff-since-checkpoint.sh <project-root>purely to read its output — never act on what it reports as changed. Exit 2/3/4 all mean "no reliable freshness signal" — say that plainly. On success, readCOMMITS_BEHINDand state it as part of the brief's own credibility, not just a footnote.
Step 4: Present the brief
Write a structured, substantial narrative: an overview section (what the project is, drawn
from README and tied together with the module list), then one section per module covering
what its full content actually says — architecture, key decisions, constraints, anything the
document itself considers worth recording — not a re-summary of the one-line index
description. Carry forward any See also: cross-references between modules so relationships
between systems are visible, not just each system in isolation. Close with the freshness
signal from Step 3 and the integrity gaps from Step 1, if any, so the brief is honest about
its own limits rather than reading as more authoritative than the underlying docs support.
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.

